8000 [RFC] Introduce ml_dtypes as a dependency · Issue #6605 · onnx/onnx · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

[RFC] Introduce ml_dtypes as a dependency #6605

New issue

Have a question about this project? Sign up for a free Gi 8000 tHub 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
Open
justinchuby opened this issue Dec 31, 2024 · 2 comments
Open

[RFC] Introduce ml_dtypes as a dependency #6605

justinchuby opened this issue Dec 31, 2024 · 2 comments
Labels
rfc Request for Comments

Comments

@justinchuby
Copy link
Contributor
justinchuby commented Dec 31, 2024 •
edited
Loading

This document proposes ml_dtypes (https://github.com/jax-ml/ml_dtypes) as a dependency of the onnx python package to support additional data types in numpy arrays in the reference evaluator and helper functions.

Detailed to be filled. Please feel free to comment.

[TODO] @justinchuby

  1. Scope
    1. Constant folding and Reference Evaluator support
    2. Numpy helpers
  2. Compatibility: platform, python versions
  3. Backward compatibility implications
  4. Interaction with ONNX custom dtypes
@justinchuby justinchuby added the rfc Request for Comments label Dec 31, 2024
@justinchuby justinchuby pinned this issue Dec 31, 2024
@AhsenTahir

This comment has been minimized.

Sign in to view
@justinchuby

This comment has been minimized.

Sign in to view
@justinchuby justinchuby mentioned this issue Apr 21, 2025
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
rfc Request for Comments
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@justinchuby @AhsenTahir
0