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

[AMP] Allow to switch whether to use promote strategy to choose kernel for O2 training. #53742

Merged
merged 7 commits into from
May 16, 2023

Conversation

Xreki
Copy link
Contributor

@Xreki Xreki commented May 12, 2023

PR types

Others

PR changes

APIs

Description

Pcard-70458

#51063 中将动态图AMP O2训练时,Kernel选择策略改成了Promote to the Widest方式。这种方式会导致一些模型在默认配置下,性能显著下降。故在auto_cast接口中添加use_promote参数来控制O2训练下的Kernel选择方法是否使用Promote策略。

@paddle-bot
Copy link

paddle-bot bot commented May 12, 2023

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.

@paddle-bot
Copy link

paddle-bot bot commented May 12, 2023

❌ The PR is not created using PR's template. You can refer to this Demo.
Please use PR's template, it helps save our maintainers' time so that more developers get helped.

Copy link
Contributor

@Ligoml Ligoml left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM for docs

@Xreki Xreki merged commit db407bf into PaddlePaddle:develop May 16, 2023
@Xreki Xreki deleted the amp/add_use_promote_dev branch May 16, 2023 05:27
Xreki added a commit to Xreki/Paddle that referenced this pull request May 16, 2023
…l for O2 training. (PaddlePaddle#53742)

* Allow to switch whether to use promote strategy to choose kernel for O2 training.

* Fix comparing error and add unittest.
lanxianghit pushed a commit that referenced this pull request May 16, 2023
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.

4 participants