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

[XPU] add conv3d, fix instance_norm, fix conv2d_transpose, test=develop, test=xpu #7642

Merged
merged 10 commits into from
Jan 10, 2022

Conversation

wangleilei001
Copy link
Contributor

add conv3d, fix instance_norm, fix conv2d_transpose, test=develop, test=xpu

@paddle-bot-old
Copy link

Thanks for your contribution!

@wangleilei001 wangleilei001 changed the title [XPU] add conv3d, fix instance_norm, fix conv2d_transpose, test=devel… [XPU] add conv3d, fix instance_norm, fix conv2d_transpose, test=develop, test=xpu Nov 16, 2021
hong19860320
hong19860320 previously approved these changes Nov 16, 2021
Copy link
Collaborator

@hong19860320 hong19860320 left a comment

Choose a reason for hiding this comment

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

LGTM

@@ -42,6 +42,9 @@ void InstanceNormCompute::Run() {
int c = param.x->dims()[1];
int height = param.x->dims()[2];
int width = param.x->dims()[3];
if (param.x->dims().size() == 5) {
Copy link
Collaborator

Choose a reason for hiding this comment

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

这部分@chenjiao review下吧

Copy link
Collaborator

@zhupengyang zhupengyang left a comment

Choose a reason for hiding this comment

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

单测是不是加下呢

lite/operators/conv3d_op.h Outdated Show resolved Hide resolved
Copy link
Collaborator

@zhupengyang zhupengyang left a comment

Choose a reason for hiding this comment

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

加下 conv_transpose 的单测

TEST(Conv_transpose, precision) {

加下 instance_norm 的单测

TEST(InstanceNorm, precision) {

lite/kernels/xpu/instance_norm_compute.cc Show resolved Hide resolved
lite/operators/conv_op.h Outdated Show resolved Hide resolved
lite/operators/conv_op.h Outdated Show resolved Hide resolved
@PaddlePaddle PaddlePaddle locked and limited conversation to collaborators Dec 20, 2021
@PaddlePaddle PaddlePaddle unlocked this conversation Dec 20, 2021
lite/kernels/xpu/instance_norm_compute.cc Outdated Show resolved Hide resolved
lite/tests/kernels/conv_transpose_compute_test.cc Outdated Show resolved Hide resolved
lite/kernels/xpu/instance_norm_compute.cc Outdated Show resolved Hide resolved
lite/kernels/xpu/instance_norm_compute.cc Outdated Show resolved Hide resolved
Copy link
Collaborator

@zhupengyang zhupengyang left a comment

Choose a reason for hiding this comment

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

conv_transpose xpu单测挂了,请关注下~

lite/tests/kernels/instance_norm_compute_test.cc Outdated Show resolved Hide resolved
@wangleilei001
Copy link
Contributor Author

conv_transpose xpu单测挂了,请关注下~
精度设置的有点高,有概率极个别值满足不了精度,已更改精度

Copy link
Collaborator

@zhupengyang zhupengyang left a comment

Choose a reason for hiding this comment

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

LGTM

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.

3 participants