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

[Paddle Inference]Add BN op TRT converter unittest #35527

Merged
merged 5 commits into from
Sep 14, 2021
Merged

[Paddle Inference]Add BN op TRT converter unittest #35527

merged 5 commits into from
Sep 14, 2021

Conversation

xiaoxiaohehe001
Copy link
Contributor

@xiaoxiaohehe001 xiaoxiaohehe001 commented Sep 7, 2021

PR types

Others

PR changes

Others

Describe

增加batch_norm op converter单测

  1. 将MomentumTensor属性在op_telller中过滤掉
  2. 待解决问题:当输入为二维时,TRT输出为四维,出现diff

@paddle-bot-old
Copy link

paddle-bot-old bot commented Sep 7, 2021

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

@xiaoxiaohehe001 xiaoxiaohehe001 changed the title add_bn_teller [Paddle Inference]Add BN op TRT converter unittest Sep 7, 2021
Comment on lines +64 to +67
"data_layout": data_layout,
"momentum": momentum,
"is_test": True,
"trainable_statistics": False
Copy link
Contributor

Choose a reason for hiding this comment

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

这几个属性在推理前向计算中应该用不到,可以不用在单测中叉乘。

Copy link
Contributor Author

Choose a reason for hiding this comment

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

data_layout、is_test、trainable_statistics在单测中必须要添加,否则会报错

@cryoco cryoco merged commit 39bc7ea into PaddlePaddle:develop Sep 14, 2021
AnnaTrainingG pushed a commit to AnnaTrainingG/Paddle that referenced this pull request Sep 29, 2021
* add_bn_

* add_bn_teller

* add_bn_teller

* add_bn_teller

* add_bn_teller
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