Bugfix code logic to allow for passing of nn.Module to TorchFXFeatureExtractor

[ORippler]

Description

• Provide a summary of the modification as well as the issue that has been resolved. List any dependencies that this modification necessitates.
  Fix code logic to allow for passing of nn.Module to TorchFXFeatureExtractor

• Fixes [Bug]: TorchFXFeatureExtractor Does not handle nn.Module passed to __init__ properly #933

Changes

• Bug fix (non-breaking change which fixes an issue)
• Refactor (non-breaking change which refactors the code base)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Checklist

• My code follows the pre-commit style and check guidelines of this project.
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing tests pass locally with my changes
• I have added a summary of my changes to the CHANGELOG (not for minor changes, docs and tests).

[samet-akcay]

Thanks @ORippler! I have a single comment only. I'll let @ashwinvaidya17 review the PR.

[ashwinvaidya17]

Thanks for spotting and addressing this! Can you add the example to the tests as well https://github.com/openvinotoolkit/anomalib/blob/main/tests/pre_merge/models/test_feature_extractor.py#L47. Just add it at the end. I'll refactor that test later on and split each torchfx tests.

[ORippler]

I would suggest changing BackboneParams.class_path in another Pull Request to be more descriptive (it can be both a path to a class and a nn.Module class-object), or split it all-together (users can either provide a class_object or a class_path to the dataclass)

[ashwinvaidya17]

I would suggest changing BackboneParams.class_path in another Pull Request to be more descriptive (it can be both a path to a class and a nn.Module class-object), or split it all-together (users can either provide a class_object or a class_path to the dataclass)

The idea was to allow jsonargparse to deserialize the the backbone class . This would allow us to use the Lightning CLI. But yeah. This whole thing needs some revisit. We are also planning on deprecating timm feature extractor. So this will change a bit

[ashwinvaidya17]

Thanks. I am fine with the changes.

[samet-akcay]

Thanks again for your contribution!

[codecov]

Codecov Report

Patch coverage: 94.73% and no project coverage change

Comparison is base (1407cfc) 86.77% compared to head (3d7f220) 86.77%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #935   +/-   ##
=======================================
  Coverage   86.77%   86.77%           
=======================================
  Files         165      165           
  Lines        6319     6322    +3     
=======================================
+ Hits         5483     5486    +3     
  Misses        836      836           

Impacted Files	Coverage Δ
...ib/models/components/feature_extractors/torchfx.py	92.98% <94.73%> (+0.38%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.