Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

test_reduce_sum_negative_axes_keepdims_exampleonnx->torch->linalg fails #653

Open
Tracked by #450
renxida opened this issue Apr 23, 2024 · 1 comment · May be fixed by llvm/torch-mlir#3218
Open
Tracked by #450

test_reduce_sum_negative_axes_keepdims_exampleonnx->torch->linalg fails #653

renxida opened this issue Apr 23, 2024 · 1 comment · May be fixed by llvm/torch-mlir#3218

Comments

@renxida
Copy link
Contributor

renxida commented Apr 23, 2024

A couple of lit tests:

  • test_reduce_sum_negative_axes_keepdims_example
  • test_reduce_sum_keepdims_example

fail to lower to linalg because:

Trying to match "(anonymous namespace)::ConvertReductionOp"
** Failure : dim argument must be a constant int list or None

https://github.com/llvm/torch-mlir/blob/3c252cdd44f411ef67e3a759319be53b46396d44/lib/Conversion/TorchToLinalg/Reduction.cpp#L453

@renxida
Copy link
Contributor Author

renxida commented Apr 23, 2024

Looks like the static shape requirement goes pretty deep through Reduction.cpp and Utils.cpp. Probably not worth it at the moment to try to support dynamic shapes. Will make the shapes static in the test cases.

@renxida renxida changed the title ReduceSum onnx->torch->linalg fails test_reduce_sum_negative_axes_keepdims_exampleonnx->torch->linalg fails Apr 23, 2024
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 a pull request may close this issue.

1 participant