Fix AdversarialPatchPyTorch compatibility with YOLO estimator

[kieranfraser]

Description

• Updating the format of targets passed to the estimator in AdversarialPatchPyTorch to reflect updates to PyTorchYolo.
• Fix for getting gradients from input tensor in PyTorchYolo.
• Adding a test in tests/estimators/object_detection/test_pytorch_yolo.py to check that adversarial patches work with the estimator.

Fixes #2148

Type of change

Please check all relevant options.

• Improvement (non-breaking)
• Bug fix (non-breaking)
• New feature (non-breaking)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Testing

Please describe the tests that you ran to verify your changes. Consider listing any relevant details of your test configuration.

• test_patch - added a test in test_pytorch_yolo.py to verify that AdversarialPatchPyTorch works with the YOLO estimator

Test Configuration:

• OS
• Python version
• ART version or commit number
• TensorFlow / Keras / PyTorch / MXNet version

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code
• 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 unit tests pass locally with my changes

[codecov-commenter]

Codecov Report

Merging #2169 (c67cd57) into dev_1.15.0 (28a3f8c) will decrease coverage by 7.56%.
The diff coverage is 50.00%.

❗ Your organization is not using the GitHub App Integration. As a result you may experience degraded service beginning May 15th. Please install the Github App Integration for your organization. Read more.

@@              Coverage Diff               @@
##           dev_1.15.0    #2169      +/-   ##
==============================================
- Coverage       85.66%   78.10%   -7.56%     
==============================================
  Files             306      306              
  Lines           27022    27026       +4     
  Branches         4962     4965       +3     
==============================================
- Hits            23148    21109    -2039     
- Misses           2602     4756    +2154     
+ Partials         1272     1161     -111     

Impacted Files	Coverage Δ
...ion/adversarial_patch/adversarial_patch_pytorch.py	57.37% <36.36%> (-7.15%)	⬇️
art/estimators/object_detection/pytorch_yolo.py	82.24% <100.00%> (+1.58%)	⬆️

... and 30 files with indirect coverage changes

[kieranfraser]

YOLO generates inconsistent loss for the same input. Fix is to freeze batch_norm and drop_out layers. However freezing these layers results in compute_loss returning NaN for some inputs due to a known issue in the python YOLO library. See #2148 for detail.

[beat-buesser]

Hi @kieranfraser Thank you very much! The changes look good to me.