Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Refactor scaler to util #3142

Merged
merged 3 commits into from
Oct 8, 2024
Merged

Refactor scaler to util #3142

merged 3 commits into from
Oct 8, 2024

Conversation

muellerzr
Copy link
Collaborator

What does this PR do?

This PR adds a new func to utils.modeling in the form of get_grad_scaler.

There's a few times when we need to repeat some of the logic for grad scalers for backward compatibility etc (see #3132) and also easier to have typos (https://github.com/huggingface/accelerate/pull/3116/files) so instead keeps one reference and refactors to its own func.

I debated on making this a class, but a functional getter similar to the one for mixed precision seemed ideal

Fixes # (issue)

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
  • 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.

@BenjaminBossan @SunMarc

@HuggingFaceDocBuilderDev
Copy link

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

BenjaminBossan
BenjaminBossan approved these changes Oct 8, 2024
View reviewed changes
Copy link
Member

@BenjaminBossan BenjaminBossan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Generally LGTM, thanks. Just one small comment and the question whether the function should also be used in benchmarks/fp8/ms_amp/ddp.py and benchmarks/fp8/ms_amp/non_distributed.py.

src/accelerate/utils/modeling.py Outdated Show resolved Hide resolved
SunMarc
SunMarc approved these changes Oct 8, 2024
View reviewed changes
Copy link
Member

@SunMarc SunMarc left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM ! +1 on @BenjaminBossan comment

@muellerzr muellerzr merged commit fb68cb9 into main Oct 8, 2024
27 of 28 checks passed
@muellerzr muellerzr deleted the refactor-scaler branch October 8, 2024 15:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@BenjaminBossan BenjaminBossan BenjaminBossan approved these changes

@SunMarc SunMarc SunMarc approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@muellerzr @HuggingFaceDocBuilderDev @BenjaminBossan @SunMarc