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

[Sparse]Optimize BatchNorm1D forward in test mode #47736

Merged
merged 13 commits into from
Nov 11, 2022

Conversation

zkh2016
Copy link
Contributor

@zkh2016 zkh2016 commented Nov 7, 2022

PR types

Performance optimization

PR changes

OPs

Describe

优化推理模式下BatchNorm1D前向性能:

  1. 当前存在重复计算inverse variance的问题,这里需要算sqrt和除法,因此效率较低
  2. 将inverse variance的计算分离除去,centerpoint推理中bn1d耗时从132ms降到26ms;整体性能提升5%。

@paddle-bot
Copy link

paddle-bot bot commented Nov 7, 2022

你的PR提交成功,感谢你对开源项目的贡献!
请关注后续CI自动化测试结果,详情请参考Paddle-CI手册
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@zkh2016 zkh2016 changed the title [Sparse]Optimize BatchNorm1D forward in test mode [WIP]Optimize BatchNorm1D forward in test mode Nov 7, 2022
@zkh2016 zkh2016 changed the title [WIP]Optimize BatchNorm1D forward in test mode [Sparse]Optimize BatchNorm1D forward in test mode Nov 9, 2022
@zkh2016 zkh2016 merged commit 6cdc18a into PaddlePaddle:develop Nov 11, 2022
zkh2016 pushed a commit to zkh2016/Paddle that referenced this pull request Nov 11, 2022
zkh2016 pushed a commit to zkh2016/Paddle that referenced this pull request Jan 31, 2023
lanxianghit pushed a commit that referenced this pull request Feb 2, 2023
cherry-pick some PR about optimize sparse kernel and fix some bug:
#47736 #47703 #47604 #46679 #48439 #49009 #49734
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