Fix Batch Size Calculation for Multi-GPU Training #5
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
![cosineai[bot]](https://avatars.githubusercontent.com/in/705948?s=60&v=4){kind=small}
Co-authored-by: Genie <genie@cosine.sh>
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Outdated suggestions cannot be applied.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.
This pull request addresses the issue of inconsistent batch size calculation during multi-GPU training. Previously, the number of batches per epoch did not correctly account for the number of GPUs, leading to an incorrect batch size. The fix involves adjusting the data parallel world size and rank when the model parallel unit (mpu) is not defined. This ensures that the number of batches is correctly calculated as the training data size divided by the number of GPUs, aligning with the expected behavior. The changes are made in the
deepspeed/runtime/engine.pyfile.Created by Genie. You can follow its reasoning on Cosine