Dev/nir

[biphasic]

No description provided.

[sheiksadique]

There is a lot of infrastructure in nirtorch. This implementation seems to not use that as of now. For sequential, this implementation is just fine. I would wait to add graph support in before merging.

[codecov]

Codecov Report

Attention: 29 lines in your changes are missing coverage. Please review.

Comparison is base (79027f3) 90.03% compared to head (336fa9d) 88.59%.
Report is 3 commits behind head on develop.

Files	Patch %	Lines
sinabs/nir.py	69.79%	29 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #104      +/-   ##
===========================================
- Coverage    90.03%   88.59%   -1.45%     
===========================================
  Files           31       32       +1     
  Lines         1244     1341      +97     
===========================================
+ Hits          1120     1188      +68     
- Misses         124      153      +29     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[biphasic]

@sheiksadique the implementation now uses nirtorch.load

[biphasic]

The current tests are not passing because the latest nirtorch version 0.2.0 has not been published, see neuromorphs/NIRTorch#6
Once it is published, we can simply re-run the CI

[github-actions]

Code Coverage Summary

Filename                                      Stmts    Miss  Cover    Missing
------------------------------------------  -------  ------  -------  --------------------------------------------------------------
sinabs/__init__.py                                8       0  100.00%
sinabs/cnnutils.py                               25      20  20.00%   13-21, 32-34, 47-56, 66-74
sinabs/conversion.py                             18       0  100.00%
sinabs/from_torch.py                             42       7  83.33%   61, 66-67, 105, 111-113
sinabs/network.py                               103      25  75.73%   42, 51, 107, 111, 151, 182, 196-197, 201-204, 222-230, 244-252
sinabs/nir.py                                    96      29  69.79%   14-22, 39-51, 74, 76, 98, 110, 114-116, 136, 140, 173, 195-212
sinabs/synopcounter.py                          142       8  94.37%   177-185, 189, 231, 273
sinabs/utils.py                                  79       4  94.94%   42, 98, 114, 164
sinabs/activation/__init__.py                     4       0  100.00%
sinabs/activation/quantize.py                    20       0  100.00%
sinabs/activation/reset_mechanism.py             18       0  100.00%
sinabs/activation/spike_generation.py            47       0  100.00%
sinabs/activation/surrogate_gradient_fn.py       46       4  91.30%   19, 23, 43, 70
sinabs/layers/__init__.py                        13       0  100.00%
sinabs/layers/alif.py                            92       5  94.57%   184, 373-374, 379, 383
sinabs/layers/channel_shift.py                   13       0  100.00%
sinabs/layers/crop2d.py                          19       2  89.47%   47-48
sinabs/layers/exp_leak.py                        25       0  100.00%
sinabs/layers/iaf.py                             43       5  88.37%   161-165
sinabs/layers/lif.py                            113       4  96.46%   139, 351-353
sinabs/layers/merge.py                           19       0  100.00%
sinabs/layers/neuromorphic_relu.py               18       1  94.44%   33
sinabs/layers/pool2d.py                          44       8  81.82%   32, 50, 58, 64, 83-91
sinabs/layers/quantize.py                        10       5  50.00%   14-15, 18-21
sinabs/layers/reshape.py                         41       2  95.12%   25, 71
sinabs/layers/stateful_layer.py                  74       2  97.30%   33, 39
sinabs/layers/to_spike.py                        45       4  91.11%   57, 82-83, 99
sinabs/layers/functional/__init__.py              2       0  100.00%
sinabs/layers/functional/alif.py                 51       0  100.00%
sinabs/layers/functional/lif.py                  53       0  100.00%
sinabs/onnx/__init__.py                           1       1  0.00%    1
sinabs/onnx/get_graph.py                         17      17  0.00%    1-31
TOTAL                                          1341     153  88.59%

Results for commit: 8fc9fd0

Minimum allowed coverage is 60%

♻️ This comment has been updated with latest results