What does this PR do?

This PR adds Perturber.grad_modifier, which allows users to define the gradient modification function.

I prefer Perturber.grad_modifier over LitModular.gradient_modifier or Adversary.gradient_modifier because

1. Defining initializer, projector and grad_modifier at one place in Perturber is a more consistent interface for users;
2. Tensor.register_hook(grad_modifier) in Perturber is always execuated, while LitModular.gradient_modifier only works in the automated optimization mode in Lighting.

I intentionally named it Perturber.grad_modifier to diffenrentiate from the in-place operations in LitModular.gradient_modifier and Adversary.gradient_modifier.

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.

• pytest
• CUDA_VISIBLE_DEVICES=0 python -m mart experiment=CIFAR10_CNN_Adv trainer=gpu trainer.precision=16 reports 70% (21 sec/epoch).
• CUDA_VISIBLE_DEVICES=0,1 python -m mart experiment=CIFAR10_CNN_Adv trainer=ddp trainer.precision=16 trainer.devices=2 model.optimizer.lr=0.2 trainer.max_steps=2925 datamodule.ims_per_batch=256 datamodule.world_size=2 reports 70% (14 sec/epoch).

Before submitting

• The title is self-explanatory and the description concisely explains the PR
• My PR does only one thing, instead of bundling different changes together
• I list all the breaking changes introduced by this pull request
• I have commented my code
• 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
• I have run pre-commit hooks with pre-commit run -a command without errors

Did you have fun?

Make sure you had fun coding 🙃