SSPCAB implementation

[djdameln]

Description

• This PR adds SSPCAB to model components and adds it as an optional extension to DRAEM.

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

[samet-akcay]

I like it! Only few minor points

[samet-akcay]

When sspcab_lambda is set to 0, it is not added to the overall loss. In this case, can we assume that sspcab_lambda=0 would be the same as enable_sspcab=false?

[samet-akcay]

if yes, it would be possible to just have a single argument, sspcab_lambda. When it is 0, it would be disabled.

[djdameln]

This would be possible, but I think it would be more clear for the user to have an explicit parameter to enable/disable SSPCAB. Especially because we would have to add some logic to enable/disable the sspcab block in the architecture of the torch model based on the value of the lambda parameter. As a user I would not expect the value of the lambda parameter to affect the architecture of the model.

[ashwinvaidya17]

Thanks!