Synchronize keys and handle missing values in dist_utils#136
Open
Jimmy-Mendez wants to merge 1 commit intoIntellindust-AI-Lab:mainfrom
Open
Synchronize keys and handle missing values in dist_utils#136Jimmy-Mendez wants to merge 1 commit intoIntellindust-AI-Lab:mainfrom
Jimmy-Mendez wants to merge 1 commit intoIntellindust-AI-Lab:mainfrom
Conversation
When training with multiple GPUs, batches with no ground truth objects cause some ranks to produce fewer loss keys (e.g., denoising losses are skipped). This results in reduce_dict attempting all_reduce on tensors of different sizes across ranks, causing a deadlock. Fix: Synchronize loss dictionary keys across all ranks before all_reduce, filling missing keys with zeros.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
When training with multiple GPUs, batches with no ground truth objects cause some ranks to produce fewer loss keys (e.g., denoising losses are skipped). This results in reduce_dict attempting all_reduce on tensors of different sizes across ranks, causing a deadlock.
Fix: Synchronize loss dictionary keys across all ranks before all_reduce, filling missing keys with zeros. (fixes #6 and fixes #34 and fixes #113 ?)