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.

Sign up for GitHub

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

Jump to bottom

Performance: Improve UpsampleBilinear forward backward performance to be on-par as oneDNN. #939

Open
fengyuan14 opened this issue Sep 26, 2024 · 2 comments
Open

Performance: Improve UpsampleBilinear forward backward performance to be on-par as oneDNN. #939

fengyuan14 opened this issue Sep 26, 2024 · 2 comments
Assignees
@majing921201
Labels
enhancement New feature or request
Milestone
PT2.6

Comments

@fengyuan14
Copy link
Contributor

🚀 The feature, motivation and pitch

Comparing with oneDNN impl, torch-xpu-ops gets gap on the performance of UpsampleBilinear forward kernel and backward kernel.

Alternatives

No response

Additional context

No response
@fengyuan14 fengyuan14 added this to the PT2.6 milestone Sep 26, 2024
@chuanqi129 chuanqi129 added the enhancement New feature or request label Oct 14, 2024
@fengyuan14
Copy link
Contributor Author

#950

@majing921201
Copy link
Contributor

majing921201 commented Oct 17, 2024 •
edited
Loading

We have implemented channels last kernel, which is on-par with cuda. But the performance still has gap with oneDNN. We will low prioritize performance optimize for oneDNN goal

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@majing921201 majing921201

Labels
enhancement New feature or request
Projects
None yet
Milestone
PT2.6
Development

No branches or pull requests
3 participants
@chuanqi129 @majing921201 @fengyuan14