[train] Deprecate torch amp wrapper utilities #51066
Merged
Conversation
This file contains 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
Signed-off-by: Justin Yu <justinvyu@anyscale.com>
Signed-off-by: Justin Yu <justinvyu@anyscale.com>
justinvyu
requested review from
hongpeng-guo,
matthewdeng,
raulchen and
woshiyyya
as code owners
justinvyu
changed the title
matthewdeng
approved these changes
Mar 4, 2025
justinvyu
changed the title
justinvyu
changed the title
hongpeng-guo
approved these changes
Mar 4, 2025
abrarsheikh
pushed a commit
that referenced
this pull request
Mar 8, 2025
The Ray Train torch AMP wrapper does not provide any value and has a worse user experience and is using deprecated APIs (`torch.cuda.amp.GradScaler`). It's easier to just have the user manage the mixed precision training themselves, as the native torch API is already pretty lightweight. Therefore, deprecate this API and its related tests. --------- Signed-off-by: Justin Yu <justinvyu@anyscale.com> Signed-off-by: Abrar Sheikh <abrar@anyscale.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Summary
The Ray Train torch AMP wrapper does not provide any value and has a worse user experience and is using deprecated APIs (
torch.cuda.amp.GradScaler). It's easier to just have the user manage the mixed precision training themselves, as the native torch API is already pretty lightweight. Therefore, deprecate this API and its related tests.As a side-node, the unit test here is testing whether or not the torch AMP API works, which is a pattern that we should avoid in the future since we don't need to duplicate tests that should already exist within torch.
Here's the migration message printed to users: