Skip to content

Conversation

@huangjiyi
Copy link
Member

PR types

Others

PR changes

Others

Describe

  • move sequence_padding from fluid to phi
@paddle-bot
Copy link

paddle-bot bot commented Feb 18, 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.

#ifdef PADDLE_WITH_XPU
template <typename T>
class UnpaddingLoDTensorFunctor<platform::XPUDeviceContext, T> {
class UnpaddingLoDTensorFunctor<paddle::platform::XPUDeviceContext, T> {
Copy link
Contributor

Choose a reason for hiding this comment

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

XPUDeviceContext替换成phi下的XPUContext

Copy link
Member Author

Choose a reason for hiding this comment

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

已处理

cpu_seq.set_lod(lod);
cpu_seq.mutable_data<T>(seq_dims, paddle::platform::CPUPlace());
auto *dev_ctx = static_cast<phi::CPUContext *>(
paddle::platform::DeviceContextPool::Instance().Get(phi::CPUPlace()));
Copy link
Contributor

Choose a reason for hiding this comment

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

这里不需要使用从全局单例里拿对象,可以试试参数传入的context行不行

Copy link
Member Author

Choose a reason for hiding this comment

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

这个改成参数传入的 context 之后 CI-Windows 上的 sequence_padding 单测没办通过,我也不太清楚原因:cry:

@huangjiyi huangjiyi force-pushed the decouple_sequence_padding branch from 8149f88 to 977527d Compare February 21, 2023 03:55
@YuanRisheng YuanRisheng merged commit 5f44360 into PaddlePaddle:develop Feb 21, 2023
@huangjiyi huangjiyi deleted the decouple_sequence_padding branch March 8, 2023 15:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

2 participants