🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/150895

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

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

❌ 1 New Failure, 1 Unrelated Failure

As of commit c02aba6 with merge base 861d2cc ():

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

This PR needs a release notes: label

If your changes are user facing and intended to be a part of release notes, please use a label starting with release notes:.

If not, please add the topic: not user facing label.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "topic: not user facing"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

I wonder to know why this revert came, I found the previous PR #145746 aimed to improve the performance and show us the runtime data. I compiled this pr and found this version seems better than its parent(commit 861d2cc). My GPU is A800, cuda version 12.0

Is that mean, the reverted PR#145746 may have something wrong in the refinement strategy?