Skip to content

Conversation

@guan404ming
Copy link
Member

@guan404ming guan404ming commented Nov 16, 2025

Related Issue

How

  • Refactored _index_tensor to handle broadcast

@guan404ming guan404ming changed the title Add decomposed operator support for interpolate [Relax][PyTorch] Add decomposed operator support for interpolate Nov 16, 2025
@guan404ming guan404ming marked this pull request as ready for review November 16, 2025 17:23
@guan404ming
Copy link
Member Author

cc @tlopex

@tlopex tlopex self-assigned this Nov 16, 2025
@guan404ming
Copy link
Member Author

I've updated the Bicubic tests since I think they're still needed. But we can also remove them if preferred.

Copy link
Member

@tlopex tlopex left a comment

Choose a reason for hiding this comment

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

LGTM! Thank you!

InterpolateNearest(), example_args, {}, expected_nearest, run_ep_decomposition=True
)
verify_model(
InterpolateBicubic(), example_args, {}, expected_bicubic, run_ep_decomposition=False
Copy link
Member

Choose a reason for hiding this comment

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

Our goal is to make every test pass when run_ep_decomposition=True , so even it has 100+ lines, you'd better rewrite the test

Copy link
Member Author

Choose a reason for hiding this comment

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

Thanks for the update.

@tlopex tlopex merged commit 6dc71f3 into apache:main Nov 19, 2025
13 checks passed
@guan404ming guan404ming deleted the add-decomposed-support-for-ip branch November 19, 2025 06:14
@guan404ming
Copy link
Member Author

Thanks!

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants