Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix tokens count for benchmark #8938

Merged
merged 1 commit into from
Aug 14, 2024

Conversation

DrownFish19
Copy link
Collaborator

PR types

Bug fixes

PR changes

Others

Description

Set total_effective_tokens=all_tokens * progress_or_epoch.
When training or fine-tuning with a max_step and using Data Parallelism (DP) or Sharding with a factor greater than 1, the calculation of total_effective_tokens may be incorrect.

Copy link

paddle-bot bot commented Aug 14, 2024

Thanks for your contribution!

@ZHUI ZHUI merged commit 8f6b001 into PaddlePaddle:develop Aug 14, 2024
4 of 10 checks passed
@DrownFish19 DrownFish19 deleted the dev_20240814_fix_benchmark branch August 14, 2024 15:21
Copy link

codecov bot commented Nov 12, 2024

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 55.04%. Comparing base (0cc8554) to head (4e867ff).
Report is 231 commits behind head on develop.

Additional details and impacted files
@@             Coverage Diff             @@
##           develop    #8938      +/-   ##
===========================================
- Coverage    55.05%   55.04%   -0.02%     
===========================================
  Files          635      635              
  Lines        99410    99436      +26     
===========================================
+ Hits         54729    54730       +1     
- Misses       44681    44706      +25     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants