fix partial state #1144
Merged
fix partial state #1144
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
|
The documentation is not available anymore as the PR was closed or merged. |
muellerzr
approved these changes
Mar 3, 2023
|
Hello, yes, here is the whole slack thread with a day's worth of debugging with respect to this: https://huggingface.slack.com/archives/C03SJ1ED7SM/p1677824690561359. In short, all FSDP integration tests were failing because the model, optimizer and lr_scheduler aren't prepared properly. |
|
Thanks, sorry for the breaking change 🙃 should've also been checking the slow tests with such a large PR before merging |
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Outdated suggestions cannot be applied.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.
What does this PR do?
PartialStateIntroduce PartialState #1055. In prepare_data_loader , state obj is created again viastate = AcceleratorState(). Now, it causes everything to go haywire,self.__dict__.update(PartialState._shared_state)reverts FSDP in AcceleratorState toMULTI_GPUpresent in PartialState. This PR fixes it.