Enable bf16 option for XLA devices#20684
Merged
sgugger merged 1 commit intohuggingface:mainfrom Dec 8, 2022
Merged
Conversation
|
The documentation is not available anymore as the PR was closed or merged. |
sgugger
reviewed
Dec 8, 2022
Collaborator
sgugger
left a comment
There was a problem hiding this comment.
I don't see how the gradient scaling path is disabled, could you share more information on that? A scaler is still defined at line 591.
Contributor
Author
Line 568 disables the code section that enables grad scaler when XLA device is detected (is_torch_tpu_available()). |
sgugger
approved these changes
Dec 8, 2022
Collaborator
sgugger
left a comment
There was a problem hiding this comment.
Ah yes, I missed the parenthesis in the condition. Thanks for explaining!
mpierrau
pushed a commit
to mpierrau/transformers
that referenced
this pull request
Dec 15, 2022
|
This PR accidentally disabled gradient scaling when using FP16 on XLA devices. |
Collaborator
|
Indeed. Do you want to make a PR with a fix @Lokiiiiii ? |
5 tasks
jeffhataws
added a commit
to jeffhataws/transformers
that referenced
this pull request
Mar 23, 2023
This PR fixes the "RuntimeError: No CUDA GPUs are available" when running with --bf16 option on Neuron. Related PRs: huggingface#20684 huggingface#22300
sgugger
pushed a commit
that referenced
this pull request
Mar 23, 2023
raghavanone
pushed a commit
to raghavanone/transformers
that referenced
this pull request
Apr 5, 2023
…ingface#22307) This PR fixes the "RuntimeError: No CUDA GPUs are available" when running with --bf16 option on Neuron. Related PRs: huggingface#20684 huggingface#22300
novice03
pushed a commit
to novice03/transformers
that referenced
this pull request
Jun 23, 2023
…ingface#22307) This PR fixes the "RuntimeError: No CUDA GPUs are available" when running with --bf16 option on Neuron. Related PRs: huggingface#20684 huggingface#22300
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do?
XLA devices like TPU and NeuronCore supports BF16 natively. This PR enables --bf16 option to work for XLA devices.
Since BF16 doesn't require gradient scaling, gradient scaling path is disabled for XLA devices.
Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
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.
@sgugger