Add onnxscript 0.4.0 as a dependency #127
Open
+22
−0
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
According to the pytorch docs (https://docs.pytorch.org/docs/stable/onnx_export.html), onnx export depends on onnxscript being installed. In PyTorch 2.9, the export behavior now defaults to torch.export instead of torch.onnx, which is a codepath that imports onnxscript eagerly. An alternative is to set dynamo=False on onnx export, but this is not recommended by PyTorch release notes. https://github.com/pytorch/pytorch/releases/tag/v2.9.0
Author
|
(note that onnxscript 0.4.0 is the version used in pytorch, see https://github.com/pytorch/pytorch/blob/8aa465f18e962fbb35eead029122cc564ed9d400/.ci/docker/common/install_onnx.sh#L23) |
Collaborator
|
Hey @Kukanani, the file you linked shows torchscript 0.5.4. Do you think it makes sense to set 0.4.0 as a minimum requirement instead of a hard constraint? |
Collaborator
|
Also, do we need a different license for this? |
Author
|
Between making the change and permalinking the code, it was updated via pytorch/pytorch#165883 to 0.5.4! So yes, that should be the correct version now. |
Author
|
onnxscript is MIT licensed, which is BSD 3-clause compatible, so I think no licensing changes are needed. |
Collaborator
|
Hi @Kukanani, I added the license and opened a PR in your fork. |
>=0.5.4 to pick up future changes, but <1.0 to prevent future breaking changes
Author
|
I merged your MR that targeted mine |
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Outdated suggestions cannot be applied.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 is a follow-up fix based on an mjlab issue: mujocolab/mjlab#230
According to the pytorch docs (https://docs.pytorch.org/docs/stable/onnx_export.html), onnx export depends on
onnxscriptbeing installed. In PyTorch 2.9, the export behavior now defaults totorch.exportinstead oftorch.onnx, which is a codepath that importsonnxscripteagerly, now requiring an explicit dependency. An alternative is to setdynamo=Falseon onnx export, but this is not recommended by PyTorch release notes (https://github.com/pytorch/pytorch/releases/tag/v2.9.0)