Add default NLP model

src/ai_modules/nlp_module/models/english_default/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "component_5_DIETClassifier.tf_model"
+all_model_checkpoint_paths: "component_5_DIETClassifier.tf_model"

src/ai_modules/nlp_module/models/english_default/component_2_LexicalSyntacticFeaturizer.feature_to_idx_dict.pkl

@@ -0,0 +1 @@
+{"-1:low": {"False": 0, "True": 1}, "-1:title": {"False": 2, "True": 3}, "-1:upper": {"False": 4, "True": 5}, "0:BOS": {"False": 6, "True": 7}, "0:EOS": {"False": 8, "True": 9}, "0:digit": {"False": 10}, "0:low": {"False": 11, "True": 12}, "0:title": {"False": 13, "True": 14}, "0:upper": {"False": 15, "True": 16}, "1:low": {"False": 17, "True": 18}, "1:title": {"False": 19, "True": 20}, "1:upper": {"False": 21, "True": 22}}

src/ai_modules/nlp_module/models/english_default/component_3_CountVectorsFeaturizer.pkl

@@ -0,0 +1 @@
+{"text": {"my": {"py/object": "numpy.intc", "dtype": "int32", "value": 37}, "name": {"py/object": "numpy.intc", "dtype": "int32", "value": 39}, "is": {"py/object": "numpy.intc", "dtype": "int32", "value": 20}, "michael": {"py/object": "numpy.intc", "dtype": "int32", "value": 32}, "eric": {"py/object": "numpy.intc", "dtype": "int32", "value": 12}, "alice": {"py/object": "numpy.intc", "dtype": "int32", "value": 2}, "am": {"py/object": "numpy.intc", "dtype": "int32", "value": 3}, "john": {"py/object": "numpy.intc", "dtype": "int32", "value": 24}, "james": {"py/object": "numpy.intc", "dtype": "int32", "value": 23}, "kristin": {"py/object": "numpy.intc", "dtype": "int32", "value": 26}, "you": {"py/object": "numpy.intc", "dtype": "int32", "value": 66}, "can": {"py/object": "numpy.intc", "dtype": "int32", "value": 8}, "call": {"py/object": "numpy.intc", "dtype": "int32", "value": 7}, "me": {"py/object": "numpy.intc", "dtype": "int32", "value": 31}, "jack": {"py/object": "numpy.intc", "dtype": "int32", "value": 22}, "peter": {"py/object": "numpy.intc", "dtype": "int32", "value": 44}, "philip": {"py/object": "numpy.intc", "dtype": "int32", "value": 45}, "the": {"py/object": "numpy.intc", "dtype": "int32", "value": 55}, "victor": {"py/object": "numpy.intc", "dtype": "int32", "value": 62}, "morgan": {"py/object": "numpy.intc", "dtype": "int32", "value": 34}, "hey": {"py/object": "numpy.intc", "dtype": "int32", "value": 17}, "dude": {"py/object": "numpy.intc", "dtype": "int32", "value": 11}, "sam": {"py/object": "numpy.intc", "dtype": "int32", "value": 49}, "mario": {"py/object": "numpy.intc", "dtype": "int32", "value": 29}, "alex": {"py/object": "numpy.intc", "dtype": "int32", "value": 1}, "molly": {"py/object": "numpy.intc", "dtype": "int32", "value": 33}, "regina": {"py/object": "numpy.intc", "dtype": "int32", "value": 46}, "yes": {"py/object": "numpy.intc", "dtype": "int32", "value": 65}, "for": {"py/object": "numpy.intc", "dtype": "int32", "value": 13}, "sure": {"py/object": "numpy.intc", "dtype": "int32", "value": 52}, "that": {"py/object": "numpy.intc", "dtype": "int32", "value": 54}, "great": {"py/object": "numpy.intc", "dtype": "int32", "value": 16}, "idea": {"py/object": "numpy.intc", "dtype": "int32", "value": 18}, "why": {"py/object": "numpy.intc", "dtype": "int32", "value": 64}, "not": {"py/object": "numpy.intc", "dtype": "int32", "value": 42}, "let": {"py/object": "numpy.intc", "dtype": "int32", "value": 28}, "do": {"py/object": "numpy.intc", "dtype": "int32", "value": 10}, "it": {"py/object": "numpy.intc", "dtype": "int32", "value": 21}, "this": {"py/object": "numpy.intc", "dtype": "int32", "value": 56}, "no": {"py/object": "numpy.intc", "dtype": "int32", "value": 40}, "thanks": {"py/object": "numpy.intc", "dtype": "int32", "value": 53}, "today": {"py/object": "numpy.intc", "dtype": "int32", "value": 59}, "maybe": {"py/object": "numpy.intc", "dtype": "int32", "value": 30}, "another": {"py/object": "numpy.intc", "dtype": "int32", "value": 4}, "day": {"py/object": "numpy.intc", "dtype": "int32", "value": 9}, "some": {"py/object": "numpy.intc", "dtype": "int32", "value": 50}, "other": {"py/object": "numpy.intc", "dtype": "int32", "value": 43}, "time": {"py/object": "numpy.intc", "dtype": "int32", "value": 57}, "nope": {"py/object": "numpy.intc", "dtype": "int32", "value": 41}, "nah": {"py/object": "numpy.intc", "dtype": "int32", "value": 38}, "interested": {"py/object": "numpy.intc", "dtype": "int32", "value": 19}, "move": {"py/object": "numpy.intc", "dtype": "int32", "value": 35}, "forward": {"py/object": "numpy.intc", "dtype": "int32", "value": 14}, "up": {"py/object": "numpy.intc", "dtype": "int32", "value": 61}, "go": {"py/object": "numpy.intc", "dtype": "int32", "value": 15}, "straight": {"py/object": "numpy.intc", "dtype": "int32", "value": 51}, "ahead": {"py/object": "numpy.intc", "dtype": "int32", "value": 0}, "moving": {"py/object": "numpy.intc", "dtype": "int32", "value": 36}, "keep": {"py/object": "numpy.intc", "dtype": "int32", "value": 25}, "walk": {"py/object": "numpy.intc", "dtype": "int32", "value": 63}, "backward": {"py/object": "numpy.intc", "dtype": "int32", "value": 6}, "back": {"py/object": "numpy.intc", "dtype": "int32", "value": 5}, "turn": {"py/object": "numpy.intc", "dtype": "int32", "value": 60}, "left": {"py/object": "numpy.intc", "dtype": "int32", "value": 27}, "rotate": {"py/object": "numpy.intc", "dtype": "int32", "value": 48}, "to": {"py/object": "numpy.intc", "dtype": "int32", "value": 58}, "right": {"py/object": "numpy.intc", "dtype": "int32", "value": 47}}, "intent": {"user_give_name": {"py/object": "numpy.intc", "dtype": "int32", "value": 1}, "user_answer_affirmative": {"py/object": "numpy.intc", "dtype": "int32", "value": 0}, "user_remain_anonymous": {"py/object": "numpy.intc", "dtype": "int32", "value": 2}, "user_say_move_forward": {"py/object": "numpy.intc", "dtype": "int32", "value": 4}, "user_say_move_backward": {"py/object": "numpy.intc", "dtype": "int32", "value": 3}, "user_say_turn_left": {"py/object": "numpy.intc", "dtype": "int32", "value": 5}, "user_say_turn_right": {"py/object": "numpy.intc", "dtype": "int32", "value": 6}}, "response": null}

src/ai_modules/nlp_module/models/english_default/component_5_DIETClassifier.entity_tag_specs.json

@@ -0,0 +1,20 @@
+[
+  {
+    "tag_name": "entity",
+    "ids_to_tags": {
+      "1": "B-person",
+      "2": "I-person",
+      "3": "L-person",
+      "4": "U-person",
+      "0": "O"
+    },
+    "tags_to_ids": {
+      "B-person": 1,
+      "I-person": 2,
+      "L-person": 3,
+      "U-person": 4,
+      "O": 0
+    },
+    "num_tags": 5
+  }
+]

src/ai_modules/nlp_module/models/english_default/component_5_DIETClassifier.index_label_id_mapping.pkl

@@ -0,0 +1 @@
+{"0": "user_answer_affirmative", "1": "user_give_name", "2": "user_remain_anonymous", "3": "user_say_move_backward", "4": "user_say_move_forward", "5": "user_say_turn_left", "6": "user_say_turn_right"}

src/ai_modules/nlp_module/models/english_default/metadata.json

@@ -0,0 +1,234 @@
+{
+    "language": "en",
+    "pipeline": [
+        {
+            "intent_tokenization_flag": false,
+            "intent_split_symbol": "_",
+            "case_sensitive": true,
+            "name": "WhitespaceTokenizer",
+            "class": "rasa.nlu.tokenizers.whitespace_tokenizer.WhitespaceTokenizer"
+        },
+        {
+            "name": "RegexFeaturizer",
+            "file": "component_1_RegexFeaturizer.pkl",
+            "class": "rasa.nlu.featurizers.sparse_featurizer.regex_featurizer.RegexFeaturizer"
+        },
+        {
+            "features": [
+                [
+                    "low",
+                    "title",
+                    "upper"
+                ],
+                [
+                    "BOS",
+                    "EOS",
+                    "low",
+                    "upper",
+                    "title",
+                    "digit"
+                ],
+                [
+                    "low",
+                    "title",
+                    "upper"
+                ]
+            ],
+            "name": "LexicalSyntacticFeaturizer",
+            "file": "component_2_LexicalSyntacticFeaturizer",
+            "class": "rasa.nlu.featurizers.sparse_featurizer.lexical_syntactic_featurizer.LexicalSyntacticFeaturizer"
+        },
+        {
+            "use_shared_vocab": false,
+            "analyzer": "word",
+            "token_pattern": "(?u)\\b\\w\\w+\\b",
+            "strip_accents": null,
+            "stop_words": null,
+            "min_df": 1,
+            "max_df": 1.0,
+            "min_ngram": 1,
+            "max_ngram": 1,
+            "max_features": null,
+            "lowercase": true,
+            "OOV_token": null,
+            "OOV_words": [],
+            "name": "CountVectorsFeaturizer",
+            "file": "component_3_CountVectorsFeaturizer.pkl",
+            "class": "rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer.CountVectorsFeaturizer"
+        },
+        {
+            "use_shared_vocab": false,
+            "analyzer": "char_wb",
+            "token_pattern": "(?u)\\b\\w\\w+\\b",
+            "strip_accents": null,
+            "stop_words": null,
+            "min_df": 1,
+            "max_df": 1.0,
+            "min_ngram": 1,
+            "max_ngram": 4,
+            "max_features": null,
+            "lowercase": true,
+            "OOV_token": null,
+            "OOV_words": [],
+            "name": "CountVectorsFeaturizer",
+            "file": "component_4_CountVectorsFeaturizer.pkl",
+            "class": "rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer.CountVectorsFeaturizer"
+        },
+        {
+            "hidden_layers_sizes": {
+                "text": [],
+                "label": []
+            },
+            "share_hidden_layers": false,
+            "transformer_size": 256,
+            "number_of_transformer_layers": 2,
+            "number_of_attention_heads": 4,
+            "use_key_relative_attention": false,
+            "use_value_relative_attention": false,
+            "max_relative_position": null,
+            "unidirectional_encoder": false,
+            "batch_size": [
+                64,
+                256
+            ],
+            "batch_strategy": "balanced",
+            "epochs": 100,
+            "random_seed": null,
+            "learning_rate": 0.001,
+            "embedding_dimension": 20,
+            "dense_dimension": {
+                "text": 512,
+                "label": 20
+            },
+            "number_of_negative_examples": 20,
+            "similarity_type": "inner",
+            "loss_type": "softmax",
+            "ranking_length": 10,
+            "maximum_positive_similarity": 0.8,
+            "maximum_negative_similarity": -0.4,
+            "use_maximum_negative_similarity": true,
+            "scale_loss": false,
+            "regularization_constant": 0.002,
+            "negative_margin_scale": 0.8,
+            "drop_rate": 0.2,
+            "drop_rate_attention": 0,
+            "weight_sparsity": 0.8,
+            "use_sparse_input_dropout": true,
+            "use_dense_input_dropout": true,
+            "evaluate_every_number_of_epochs": 20,
+            "evaluate_on_number_of_examples": 0,
+            "intent_classification": true,
+            "entity_recognition": true,
+            "use_masked_language_model": false,
+            "BILOU_flag": true,
+            "tensorboard_log_directory": null,
+            "tensorboard_log_level": "epoch",
+            "name": "DIETClassifier",
+            "file": "component_5_DIETClassifier",
+            "class": "rasa.nlu.classifiers.diet_classifier.DIETClassifier"
+        },
+        {
+            "BILOU_flag": true,
+            "features": [
+                [
+                    "low",
+                    "title",
+                    "upper"
+                ],
+                [
+                    "low",
+                    "bias",
+                    "prefix5",
+                    "prefix2",
+                    "suffix5",
+                    "suffix3",
+                    "suffix2",
+                    "upper",
+                    "title",
+                    "digit",
+                    "pattern"
+                ],
+                [
+                    "low",
+                    "title",
+                    "upper"
+                ]
+            ],
+            "max_iterations": 50,
+            "L1_c": 0.1,
+            "L2_c": 0.1,
+            "name": "CRFEntityExtractor",
+            "files": {
+                "entity": "component_6_CRFEntityExtractor.entity.pkl"
+            },
+            "class": "rasa.nlu.extractors.crf_entity_extractor.CRFEntityExtractor"
+        },
+        {
+            "name": "EntitySynonymMapper",
+            "file": null,
+            "class": "rasa.nlu.extractors.entity_synonyms.EntitySynonymMapper"
+        },
+        {
+            "hidden_layers_sizes": {
+                "text": [
+                    256,
+                    128
+                ],
+                "label": [
+                    256,
+                    128
+                ]
+            },
+            "share_hidden_layers": false,
+            "transformer_size": null,
+            "number_of_transformer_layers": 0,
+            "number_of_attention_heads": 4,
+            "use_key_relative_attention": false,
+            "use_value_relative_attention": false,
+            "max_relative_position": null,
+            "unidirectional_encoder": false,
+            "batch_size": [
+                64,
+                256
+            ],
+            "batch_strategy": "balanced",
+            "epochs": 100,
+            "random_seed": null,
+            "learning_rate": 0.001,
+            "embedding_dimension": 20,
+            "dense_dimension": {
+                "text": 512,
+                "label": 512
+            },
+            "number_of_negative_examples": 20,
+            "similarity_type": "inner",
+            "loss_type": "softmax",
+            "ranking_length": 10,
+            "maximum_positive_similarity": 0.8,
+            "maximum_negative_similarity": -0.4,
+            "use_maximum_negative_similarity": true,
+            "scale_loss": true,
+            "regularization_constant": 0.002,
+            "weight_sparsity": 0.0,
+            "negative_margin_scale": 0.8,
+            "drop_rate": 0.2,
+            "drop_rate_attention": 0,
+            "use_sparse_input_dropout": false,
+            "use_dense_input_dropout": false,
+            "evaluate_every_number_of_epochs": 20,
+            "evaluate_on_number_of_examples": 0,
+            "use_masked_language_model": false,
+            "retrieval_intent": null,
+            "tensorboard_log_directory": null,
+            "tensorboard_log_level": "epoch",
+            "name": "ResponseSelector",
+            "intent_classification": true,
+            "entity_recognition": false,
+            "BILOU_flag": null,
+            "file": null,
+            "class": "rasa.nlu.selectors.response_selector.ResponseSelector"
+        }
+    ],
+    "trained_at": "20201211-201842",
+    "rasa_version": "1.10.18"
+}