Skip to content

Conversation

@lw
Copy link

@lw lw commented Dec 11, 2025

The data-dependent check on the standard deviation (in case any of its elements is zero) caused a graph break when using torch.compile. Instead this can be replaced by an in-graph assert, which can even become a on-device assert in order to support CUDA graphs.

The data-dependent check on the standard deviation (in case any of its elements is zero) caused a graph break when using torch.compile. Instead this can be replaced by an in-graph assert, which can even become a on-device assert in order to support CUDA graphs.
@pytorch-bot
Copy link

pytorch-bot bot commented Dec 11, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/9314

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

❌ 27 New Failures

As of commit 89c1fd6 with merge base 96e7797 (image):

NEW FAILURES - The following jobs have failed:

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

@meta-cla meta-cla bot added the cla signed label Dec 11, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants