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

Replace clip, bce_loss, full and full_like with elementwise #39197

Merged

Conversation

AnnaTrainingG
Copy link
Contributor

@AnnaTrainingG AnnaTrainingG commented Jan 25, 2022

PR types

Others

PR changes

OPs

Describe

Replace clip, bce_loss, full and full_like kernel with elementwiseKernel
op benchmark 异常说明:
"prod_2 (backward)" ,"prod_2 (forward)" ,"sum_2 (forward)" ,"sum_2 (backward)" 提示性能下降均与本次修改无关。
本地性能测试结果如下:

op name axise dtype shape dev (us) new (us)  
sum [1] float32 [16, 8, 128] 1.546 1.546 1.00
sum backward [1] float32 [16, 8, 128] 2.783 2.83 1.02
prod [1] float32 [16, 8, 128] 1.56 1.599 1.03
prod backward [1] float32 [16, 8, 128] 4.496 4.491 1.00

case 比较小 机器波动影响比较大。

@paddle-bot-old
Copy link

Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

xingfeng01
xingfeng01 previously approved these changes Jan 25, 2022
@paddle-bot-old
Copy link

paddle-bot-old bot commented Feb 3, 2022

Sorry to inform you that ec98ea6's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually.

Copy link
Contributor

@Xreki Xreki 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 op benchmark ci

Copy link
Contributor

@ZzSean ZzSean 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 ci op benchmark

@AnnaTrainingG AnnaTrainingG merged commit 424700f into PaddlePaddle:develop Feb 8, 2022
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.

5 participants