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

Remove NumPy <2 pin #1375

Merged
merged 3 commits into from
Aug 23, 2024
Merged

Remove NumPy <2 pin #1375

merged 3 commits into from
Aug 23, 2024

Conversation

seberg
Copy link
Contributor

@seberg seberg commented Aug 19, 2024

This PR removes the NumPy<2 pin which is expected to work for
RAPIDS projects once CuPy 13.3.0 is released (CuPy 13.2.0 had
some issues preventing the use with NumPy 2).

@seberg seberg added non-breaking Non-breaking change improvement Improvement / enhancement to an existing function labels Aug 19, 2024
@github-actions github-actions bot added the conda conda issue label Aug 19, 2024
@jakirkham jakirkham marked this pull request as ready for review August 22, 2024 23:10
@jakirkham jakirkham requested a review from a team as a code owner August 22, 2024 23:10
@jakirkham jakirkham requested a review from bdice August 22, 2024 23:10
@jakirkham
Copy link
Member

Closing and reopening to rerun CI with CuPy 13.3.0 (since it was released earlier today)

@jakirkham jakirkham closed this Aug 22, 2024
@jakirkham jakirkham reopened this Aug 22, 2024
@jakirkham
Copy link
Member

It is worth noting that there is a cycle in the dependency graph between cuDF requiring Dask-CUDA in its tests and Dask-CUDA requiring cuDF

Unfortunately that means to test either one with NumPy 2, we need to rebuild both with NumPy 2 first. Since it is not possible to rebuild both and test at the same time, we need to pick one to rebuild first so we can test the other

Given Dask-CUDA's library usage of NumPy 2 is relatively simple (wrapping memory and very minor adjustments), rebuilding Dask-CUDA first seems like the safer path. We can then rebuild and test cuDF with NumPy 2 and Dask-CUDA. Once cuDF packages are out we can come back and test Dask-CUDA on NumPy 2 (using a no-op PR to run CI tests and spot checking CI logs for NumPy 2 usage with cuDF)

Leaving this note for completeness about this plan. Will also link the no-op testing PR here

@jakirkham
Copy link
Member

Should add, ran a quick sanity check using Ruff's NumPy 2 plugin and it came back clean

% ruff check dask_cuda --select NPY201
All checks passed!

So hopefully that provides us some more confidence (in addition to repeated offline testing done as part of NumPy 2 bringup, CuPy 13.3.0 development, etc.)

Copy link
Member

@jakirkham jakirkham left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks Sebastian! 🙏

@jakirkham
Copy link
Member

/merge

@rapids-bot rapids-bot bot merged commit 4a02fcc into rapidsai:branch-24.10 Aug 23, 2024
23 checks passed
@jakirkham
Copy link
Member

Will use PR ( #1378 ) for testing

@seberg seberg deleted the my_new_branch branch August 23, 2024 08:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
conda conda issue improvement Improvement / enhancement to an existing function non-breaking Non-breaking change
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants