Add Gradio

[julien-blanchon]

Add Simple Gladio support

Add new tool inference_gladio.py for gladio integration

optional arguments:
  -h, --help            show this help message and exit
  --config CONFIG       Path to a model config file
  --weight_path WEIGHT_PATH
                        Path to a model weights
  --threshold THRESHOLD
                        Value to threshold anomaly scores into 0-100 range

Example

python tools/inference_gradio.py --config ./anomalib/models/padim/config.yaml --weight_path ./results/padim/mvtec/bottle/weights/model.ckpt

Changes

• Add Gladio Support
• Add gradio>=2.9.4 dependencies
• 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

[julien-blanchon]

#275

[ashwinvaidya17]

Thanks for this contribution! I have a minor comment. This would help with the checks: pre-commit https://openvinotoolkit.github.io/anomalib/guides/using_pre_commit.html

[samet-akcay]

Thanks @julien-blanchon! This is a great addition! 🔥

I only have a few comments, hopefully wouldn't require so much of your time. If yes, we could do those changes as well. Thanks!

[julien-blanchon]

I will make a new commit this tomorrow with some patch

[julien-blanchon]

I've just push a typo fix, feel free to reformat get_inferencer and to move it as an utils.

[julien-blanchon]

The infer function may also be moved outside and reused.

Maybe a better description and interface could be done.

[samet-akcay]

The infer function may also be moved outside and reused.
Maybe a better description and interface could be done.

@julien-blanchon thanks a lot! Yeah, I agree, we could do another round of refactoring to re-use some of these across other inference stuff. This is the scope of another PR that we could create in the future, though.

Really appreciate your efforts here. Thanks again!