Skip to content

Conversation

@zhiqiu
Copy link
Contributor

@zhiqiu zhiqiu commented Sep 3, 2021

PR types

New features

PR changes

OPs

Describe

add clip_by_norm fp16 kernel

@paddle-bot-old
Copy link

paddle-bot-old bot commented Sep 3, 2021

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

@zhiqiu zhiqiu requested a review from zhangting2020 September 3, 2021 11:17
@AnnaTrainingG
Copy link
Contributor

AnnaTrainingG commented Sep 6, 2021

LGTM for TensorReduceFunctorImpl

Copy link
Contributor

@Aurelius84 Aurelius84 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 DataType Registarar.

Copy link
Contributor

@chenwhql chenwhql 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 const_cast

// merge ids in selected rows first
math::scatter::MergeAdd<platform::CUDADeviceContext, platform::float16>
merge_func;
SelectedRows* merged_input =
Copy link
Contributor

Choose a reason for hiding this comment

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

这里不太理解为什么要将临时的merged_input放到scope中,看起来只是在kernel中使用的临时变量

Copy link
Contributor Author

Choose a reason for hiding this comment

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

嗯嗯,后面再优化下这部分代码

@zhiqiu zhiqiu merged commit 7aa4d87 into PaddlePaddle:develop Sep 8, 2021
2742195759 pushed a commit to 2742195759/Paddle that referenced this pull request Sep 10, 2021
* add clip_by_norm fp16 kernel

* add ut
AnnaTrainingG pushed a commit to AnnaTrainingG/Paddle that referenced this pull request Sep 29, 2021
* add clip_by_norm fp16 kernel

* add ut
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