Check if automicrobatching is on before using FSDP modules in Eval #3521
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
This reverts commit ae598f9.
mvpatel2000
reviewed
Aug 5, 2024
composer/trainer/trainer.py
Outdated
Comment on lines
3756
to
3760
| if self.state.auto_microbatching: | ||
| sync_hook = _create_sync_hook(self.state) | ||
| if self.state.fsdp_enabled and len(self.automicrobatch_fsdp_hook_handles) == 0: | ||
| self.automicrobatch_fsdp_hook_handles = _readd_fsdp_sync_hooks(self.fsdp_modules, sync_hook) | ||
| self.num_consecutive_non_OOM_batches = 0 |
There was a problem hiding this comment.
I don't think this solves the issue -- if you resume with automicrobatching on, then it will still fail. Instead, the top level if statement should also check if attr fsdp_modules exists.
Can you verify also this works with Jacob's run?
There was a problem hiding this comment.
I also don't understand -- shouldn't self.fsdp_modules be set during FSDP wrap on init?
There was a problem hiding this comment.
yea it doesnt the issue is in checkpointing, where prepare_fsdp_module is called and doesnt return anything
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Outdated suggestions cannot be applied.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.
Runs resuming from pre-sync-dropping Composer broke due to FSDP modules not existing in the Eval Loop. This checks for automicrobatching before calling
self.fsdp_modules.