[Relax][PyTorch] Add UpSample Bicubic Op Support for Exported Program and FX graph #17932
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Hzfengsy
approved these changes
May 10, 2025
Member
Hzfengsy
left a comment
Hzfengsy
left a comment
There was a problem hiding this comment.
Overall LGTM, but a minor question
ShiboXing
pushed a commit
to ShiboXing/tvm
that referenced
this pull request
Aug 10, 2025
… and FX graph (apache#17932) * add upsample bicubic op support into torch frontend * fix cubic alpha value for all interpolate func * fix cubic alpha values in all test script * update the mapping code in frontend * fix lint issue --------- Co-authored-by: deivanayakisankaralingam <deiva@Deivanayaki>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR introduces support for the upsample_bicubic2d operation in PyTorch, expanding the existing upsampling functionalities to include bicubic interpolation. The implementation aligns with the default behavior of PyTorch's torch.nn.functional.interpolate function, which utilizes a cubic convolution kernel with an alpha value of -0.75. Notably, the default alpha value has been adjusted from -0.5 to -0.75 to match the exact results produced by PyTorch's implementation. By adding this support, the following models are now able to run successfully.