support bf16

[etemadiamd]

What does this PR do?

Added bf16 support.

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• [ x] Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

[etemadiamd]

@mfuntowicz @amyeroberts Started a new PR for bf16 support here!
CC: @AdrianAbeyta @amathews-amd

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.

[LysandreJik]

WDYT @ydshieh ?

The torch.cuda.is_bf16_supported command was introduced in PyTorch 1.10. As I believe we dropped support for torch 1.9, wdyt of this change?

[ydshieh]

+1 for this change ❤️ .

[ydshieh]

Hi @etemadiamd I think your PR is on top of a commit on main that is old. Could you pull the latest changes to your local main and rebase your PR branch on top of main? This should fix several failing tests.

[ydshieh]

need a make fixup (or simply make style and make quality) to pass CI

[etemadiamd]

Hi @ydshieh, Thanks! Applied "make style and make quality". The "ci/circleci: check_code_quality" and "setup_and_quality" tests passed the CI checks. Now "run_tests" and "ci/circleci: examples_torch" tests failed with "Failed: Timeout >120.0s". Any idea to fix the failures?

[ydshieh]

Thanks for the fix! For the timeout one, we can ignore it.

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint.

[amyeroberts]

Thanks for this improvement!

[ydshieh]

Thanks again @etemadiamd !