🐞 Fix inference for draem

[djdameln]

Description

This PR fixes inference for DRAEM and possibly other models.

• The transform config files were not loaded correctly in each of the three inferencers, so the default transform config was always used. This is solved now by retrieving the transform config from the config file using the correct key.
• The torch model from the DRAEM implementation now returns an anomaly map when in inference mode, instead of the confidence scores for both classes. This makes the model compatible with the torch and openvino inferencers.

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

[ashwinvaidya17]

If OpenVINO inference works for DRAEM can you add it to this test https://github.com/openvinotoolkit/anomalib/blob/main/tests/pre_merge/deploy/test_inferencer.py#L88

[samet-akcay]

If OpenVINO inference works for DRAEM can you add it to this test https://github.com/openvinotoolkit/anomalib/blob/main/tests/pre_merge/deploy/test_inferencer.py#L88

I agree, it might be an idea to add DRAEM to the tests.

[samet-akcay]

Thanks for the fix!

[ashwinvaidya17]

Thanks for identifying the issue!