fix(buffer): reshuffle with correct epoch in ShuffleSelector#593
Merged
Conversation
Collaborator
Author
|
/unittest-diff |
unittest: Run #1786
🎉 All tests passed!Github Test Reporter by CTRF 💚 |
pan-x-c
approved these changes
Jun 26, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
[Please describe the background, purpose, changes made, and how to test this PR]
Background
ShuffleSelector regenerates its shuffle order via _get_orders(), which derives the epoch from current_index // dataset_size. Because current_index was incremented after the reshuffle, the epoch number lagged by one.
Purpose
Fix an off-by-one bug where consecutive epochs reused the same permutation (epoch 0 == epoch 1), and checkpoint resume diverged from a continuous run.
Changes
How to test
pytest tests/buffer/task_scheduler_test.py -q
Verified on DSW (env trinity): 9 passed; the new test fails on the pre-fix code (epoch0 == epoch1) and passes after the fix.
Checklist
Please check the following items before code is ready to be reviewed.