Fix AttributeError in hf2megads weight converter #483
Conversation
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
Add DummyOptimizerWithStateDict class that inherits from torch.optim.Optimizer to resolve the AttributeError that occurs when DeepSpeed tries to save checkpoints during weight conversion. The built-in DummyOptim created when passing optimizer=None to deepspeed.initialize() lacks a state_dict() method. Additionally, DeepSpeed validates that the optimizer is an instance of expected types (Optimizer, None, or Callable). This fix provides a custom optimizer that: - Inherits from torch.optim.Optimizer to pass DeepSpeed's type check - Implements required methods: step(), state_dict(), and load_state_dict() - Returns empty state since optimizer state is not needed during conversion Co-authored-by: aravneelaws <[email protected]>
KeitaW
requested review from
GuanhuaWang,
jeffra and
tjruwase
as code owners
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
This PR is aiming to address #482 by adding
DummyOptimizerWithStateDictclass that inherits fromtorch.optim.Optimizerto resolve the AttributeError that occurs when DeepSpeed tries to save checkpoints during weight conversion.The built-in DummyOptim created when passing
optimizer=Nonetodeepspeed.initialize()lacks astate_dict()method. Additionally, DeepSpeed validates that the optimizer is an instance of expected types (Optimizer, None, or Callable).This fix provides a custom optimizer that:
torch.optim.Optimizerto pass DeepSpeed's type checkstep(),state_dict(), andload_state_dict()