perf(rollout): compress loss masks with run-length encoding #1473
+169
−9
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
Reduce memory usage and network overhead by compressing binary loss masks using run-length encoding (RLE) when transferring data between rollout and training components. - Add compress_loss_mask and decompress_loss_mask utilities - Compress masks in RolloutManager before sending - Decompress in FSDP and Megatron data processing - Add comprehensive tests for RLE functions
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Outdated suggestions cannot be applied.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.
There is a TODO in rollout.py to add loss mask compression. I think it makes sense as high compression rate is expected.
Detailed actions: