You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by "Deivanayaki-S (via GitHub)" <gi...@apache.org> on 2025/05/09 09:40:36 UTC
[PR] [Relax][PyTorch] Add UpSample Bicubic Op Support for Exported Program and FX graph [tvm]
Deivanayaki-S opened a new pull request, #17932:
URL: https://github.com/apache/tvm/pull/17932
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.
1. hustvl/yolos-tiny
2. depth-anything/Depth-Anything-V2-Small-hf
3. yolo_wgisd
4. facebook/dinov2-base
5. OpenGVLab/InternViT-300M-448px
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [Relax][PyTorch] Add UpSample Bicubic Op Support for Exported Program and FX graph [tvm]
Posted by "Deivanayaki-S (via GitHub)" <gi...@apache.org>.
Deivanayaki-S commented on code in PR #17932:
URL: https://github.com/apache/tvm/pull/17932#discussion_r2082841565
##########
tests/python/relax/test_frontend_from_fx.py:
##########
@@ -3250,7 +3250,7 @@ def main(
method="nearest_neighbor",
coordinate_transformation_mode="asymmetric",
rounding_method="round",
- cubic_alpha=-0.5,
+ cubic_alpha=-0.75,
Review Comment:
@Hzfengsy Since the default value of cubic_alpha has changed, the corresponding test scripts for other operations may need to be updated to reflect this change.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [Relax][PyTorch] Add UpSample Bicubic Op Support for Exported Program and FX graph [tvm]
Posted by "Hzfengsy (via GitHub)" <gi...@apache.org>.
Hzfengsy commented on code in PR #17932:
URL: https://github.com/apache/tvm/pull/17932#discussion_r2082811358
##########
tests/python/relax/test_frontend_from_fx.py:
##########
@@ -3250,7 +3250,7 @@ def main(
method="nearest_neighbor",
coordinate_transformation_mode="asymmetric",
rounding_method="round",
- cubic_alpha=-0.5,
+ cubic_alpha=-0.75,
Review Comment:
Why need to change the existing tests?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [Relax][PyTorch] Add UpSample Bicubic Op Support for Exported Program and FX graph [tvm]
Posted by "Hzfengsy (via GitHub)" <gi...@apache.org>.
Hzfengsy merged PR #17932:
URL: https://github.com/apache/tvm/pull/17932
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org