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.

Sign up for GitHub

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

Jump to bottom

Align ReduceMin / ReduceMax etc. handling of empty tensors with spec #341

Open
robertknight opened this issue Sep 1, 2024 · 1 comment
Open

Align ReduceMin / ReduceMax etc. handling of empty tensors with spec #341

robertknight opened this issue Sep 1, 2024 · 1 comment
Labels
Spec compliance Issues with RTen behavior not matching the ONNX specifications

Comments

@robertknight
Copy link
Owner

The rten implementation of ReduceMin and ReduceMax returns an error if the tensor is empty. However the spec for eg. ReduceMax says:

Reduction over an empty set of values yields minus infinity (if supported by the datatype) or the minimum value of the data type otherwise.
@robertknight robertknight added the Spec compliance Issues with RTen behavior not matching the ONNX specifications label Sep 1, 2024
@robertknight
Copy link
Owner Author

This came up in the context of the DynamicQuantizeLinear operator, which uses ReduceMin / ReduceMax in its definition. I noticed that there was no explicit handling of the case where the input is empty.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
Spec compliance Issues with RTen behavior not matching the ONNX specifications
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@robertknight