Skip to content

prefetch data during training#2534

Merged
wanghan-iapcm merged 2 commits intodeepmodeling:develfrom
njzjz:parallel-data-train
May 19, 2023
Merged

prefetch data during training#2534
wanghan-iapcm merged 2 commits intodeepmodeling:develfrom
njzjz:parallel-data-train

Conversation

@njzjz
Copy link
Member

@njzjz njzjz commented May 17, 2023

Fix #2229.

Train models and prefetch data in parallel to decouple the time when data is produced from the time when data is consumed.

njzjz added 2 commits May 16, 2023 19:29
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
@codecov
Copy link

codecov bot commented May 17, 2023

Codecov Report

Patch coverage: 92.30% and project coverage change: +0.06 🎉

Comparison is base (234aaca) 74.98% compared to head (f2098d4) 75.05%.

Additional details and impacted files
@@            Coverage Diff             @@
##            devel    #2534      +/-   ##
==========================================
+ Coverage   74.98%   75.05%   +0.06%     
==========================================
  Files         228      228              
  Lines       23882    23927      +45     
  Branches     1680     1666      -14     
==========================================
+ Hits        17909    17959      +50     
+ Misses       4899     4893       -6     
- Partials     1074     1075       +1     
Impacted Files Coverage Δ
deepmd/train/trainer.py 82.92% <92.30%> (+0.68%) ⬆️

... and 2 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

@wanghan-iapcm wanghan-iapcm requested a review from iProzd May 17, 2023 05:57
@wanghan-iapcm wanghan-iapcm merged commit f6d883f into deepmodeling:devel May 19, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[BUG] Training wall time is abnormally long when sets contain many systems

3 participants