Skip to content

Support ceil_mode=True on maxpool2d in XNNPACK delegate #7355

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

Merged
merged 1 commit into from
Dec 21, 2024

Conversation

GregoryComer
Copy link
Member

Summary: Support ceil_mode=True for MaxPool2d in XNNPACK delegate by conditionally updating padding. This works only when the input shape is static, as it requires statically computing the change in padding. As such, the partitioner constraints are also updated to reflect this.

Differential Revision: D67386151

Copy link

pytorch-bot bot commented Dec 18, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7355

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit f0474f4 with merge base 6c3a792 (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Dec 18, 2024
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D67386151

@facebook-github-bot
Copy link
Contributor

@GregoryComer has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

Copy link
Contributor

@mcr229 mcr229 left a comment

Choose a reason for hiding this comment

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

This is smart, i like it

GregoryComer added a commit to GregoryComer/executorch that referenced this pull request Dec 19, 2024
Summary:
Support ceil_mode=True for MaxPool2d in XNNPACK delegate by conditionally updating padding. This works only when the input shape is static, as it requires statically computing the change in padding. As such, the partitioner constraints are also updated to reflect this.


Test Plan: Added a new test (test_fp32_maxpool2d_ceilmode) that exercises various permutations of input with ceil_mode=True.

Reviewed By: digantdesai, mcr229

Differential Revision: D67386151

Pulled By: GregoryComer
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D67386151

GregoryComer added a commit to GregoryComer/executorch that referenced this pull request Dec 20, 2024
Summary:
Support ceil_mode=True for MaxPool2d in XNNPACK delegate by conditionally updating padding. This works only when the input shape is static, as it requires statically computing the change in padding. As such, the partitioner constraints are also updated to reflect this.


Test Plan: Added a new test (test_fp32_maxpool2d_ceilmode) that exercises various permutations of input with ceil_mode=True.

Reviewed By: digantdesai, mcr229

Differential Revision: D67386151

Pulled By: GregoryComer
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D67386151

Summary:
Support ceil_mode=True for MaxPool2d in XNNPACK delegate by conditionally updating padding. This works only when the input shape is static, as it requires statically computing the change in padding. As such, the partitioner constraints are also updated to reflect this.


Test Plan: Added a new test (test_fp32_maxpool2d_ceilmode) that exercises various permutations of input with ceil_mode=True.

Reviewed By: digantdesai, mcr229

Differential Revision: D67386151

Pulled By: GregoryComer
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D67386151

@facebook-github-bot facebook-github-bot merged commit 82763a9 into pytorch:main Dec 21, 2024
45 of 46 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported release notes: backends [DO NOT USE] Changes to any of the backend delegates
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants