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

[2/4] CUDNNv8 ResNet Fusion: Add fused_scale_bias_add_relu OP #58504

Merged

Conversation

Tom-Zheng
Copy link
Contributor

PR types

New features

PR changes

OPs

Description

This PR adds fused_scale_bias_add_relu op, which is part of the CUDNNv8 ResNet fusion integration.

@Tom-Zheng
Copy link
Contributor Author

@Xreki This PR is similar to #55026, would you please take a look?

@paddle-bot paddle-bot bot added the contributor External developers label Oct 30, 2023
@Tom-Zheng
Copy link
Contributor Author

PR-CI-OP-benchmark and PR-CI-Coverage need manual approval and should not block review process.

@onecatcn onecatcn removed the request for review from Xreki November 2, 2023 02:45
Copy link

paddle-ci-bot bot commented Nov 8, 2023

Sorry to inform you that 09f5aff's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually.

@Tom-Zheng Tom-Zheng force-pushed the gh_add_fused_scale_bias_add_relu branch 2 times, most recently from 8af360a to 99c5d30 Compare November 9, 2023 08:55
@Tom-Zheng
Copy link
Contributor Author

@phlrain would you please approve check_dygraph=False? This OP is only intended for static graph IR pass.
@luotao1 would you please approve use of skip_check_grad_ci; use of skipIf ; setting atol, rtol; use of const_cast?

- op : fused_scale_bias_add_relu
args : (Tensor x1, Tensor scale1, Tensor bias1, Tensor x2, Tensor scale2, Tensor bias2, bool fuse_dual, bool exhaustive_search)
optional : scale2, bias2
output : Tensor(y)
Copy link
Contributor

Choose a reason for hiding this comment

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

一般来说,单输出都取名为out

Copy link
Contributor Author

Choose a reason for hiding this comment

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

会在下个PR修改.

fuse_dual));
}
// set output dims
y->set_dims(x1.dims());
Copy link
Contributor

Choose a reason for hiding this comment

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

严格来说,这里也需要设置dtype和layout

Copy link
Contributor Author

Choose a reason for hiding this comment

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

会在下个PR修改.

Copy link
Collaborator

@phlrain phlrain 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 check_dygaph =False

@Tom-Zheng
Copy link
Contributor Author

@zyfncg Hi, would you please merge?

@zyfncg zyfncg merged commit df31e4c into PaddlePaddle:develop Nov 13, 2023
28 checks passed
danleifeng pushed a commit to danleifeng/Paddle that referenced this pull request Nov 14, 2023
SecretXV pushed a commit to SecretXV/Paddle that referenced this pull request Nov 28, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
contributor External developers NVIDIA
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants