Implement a portable adaptive_avgpool2d kernel#18067
Implement a portable adaptive_avgpool2d kernel#18067meta-codesync[bot] merged 1 commit intopytorch:mainfrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/18067
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 5 Pending, 4 Unrelated FailuresAs of commit 2576b9b with merge base be0cdee ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
BROKEN TRUNK - The following jobs failed but was present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
e2b42a9 to
bb23c30
Compare
|
@GregoryComer has imported this pull request. If you are a Meta employee, you can view this in D95997748. |
bb23c30 to
9fdda5d
Compare
| # See the README.md file in this directory for a description of the syntax used | ||
| # by this file. | ||
|
|
||
| - op: _adaptive_avg_pool2d.out |
There was a problem hiding this comment.
why do we need a underscore at the front? is this kernel internal only and won;t expose to user?
There was a problem hiding this comment.
Good question. This is actually the ATen op name, so this is referencing the official schema here: https://github.com/pytorch/pytorch/blob/main/aten/src/ATen/native/native_functions.yaml#L12443. I'm not sure why the ATen op itself starts with an underscore.
9fdda5d to
2576b9b
Compare
Differential Revision: D95997748 Pull Request resolved: pytorch#18067
Summary
Add a basic _adaptive_avgpool2d kernel to portable. It shows up in some CV models.
cc @larryliu0820 @manuelcandales