Update tests/models/*.py to use devices instead of gpus or ipus

[mathemusician]

What does this PR do?

Update test to deprecate num_processes, gpus, tpu_cores, and ipus from the Trainer constructor

Part of Pull 11040.

🌳 Trello Board Link

Does your PR introduce any breaking changes? If yes, please list them.

Before submitting

• Was this discussed/approved via a GitHub issue? (not for typos and docs)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure your PR does only one thing, instead of bundling different changes together?
• Did you make sure to update the documentation with your changes? (if necessary)
• Did you write any new necessary tests? (not for typos and docs)
• Did you verify new and existing tests pass locally with your changes?
• Did you list all the breaking changes introduced by this pull request?
• Did you update the CHANGELOG? (not for typos, docs, test updates, or internal minor changes/refactorings)

PR review

Anyone in the community is welcome to review the PR.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:

• Is this pull request ready for review? (if not, please submit in draft mode)
• Check that all items from Before submitting are resolved
• Make sure the title is self-explanatory and the description concisely explains the PR
• Add labels and milestones (and optionally projects) to the PR so it can be classified

Did you have fun?

Make sure you had fun coding 🙃

[mathemusician]

A lot of the tests in test_gpu.py don't work anymore because the logic is completely different when using devices instead of gpu. Please let me know if yall think we should tweak or sunset those tests.

[mathemusician]

Also, it may be apparent from test_hooks.py that the Trainer constructor may need to be tweaked to get

called == expected

I didn't want to implement any major changes until I got input from yall

[mathemusician]

I deleted a few tests from tests/models/test_gpu.py because they test the behavior of gpus=0 or gpus=None. This would test accelerator=cpu behavior, not accelerator=gpu, and therefore do not belong in test_gpu.py.

[mathemusician]

Also, it may be apparent from test_hooks.py that the Trainer constructor may need to be tweaked to get

called == expected

I didn't want to implement any major changes until I got input from yall

Nvm, figured this out

[mathemusician]

A lot of the tests in test_gpu.py don't work anymore because the logic is completely different when using devices instead of gpu. Please let me know if yall think we should tweak or sunset those tests.

Decided to sunset some tests.

[mathemusician]

tensorboard is failing?

[daniellepintz]

@mathemusician that is a known issue not related to your PR: pytorch/pytorch#69904

[mathemusician]

Thanks @rohitgr7! Those TPU tests took me on a roller coaster ride to nowhere.