Skip to content

DOC Update array API common checks docs #31050

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

Merged
merged 4 commits into from
May 13, 2025
Merged

DOC Update array API common checks docs #31050

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
a5c67a6
Update array API common checks docs
betatim Mar 21, 2025
25de158
Update doc/modules/array_api.rst
betatim May 12, 2025
5123908
Add newline
betatim May 12, 2025
58bd8df
Merge branch 'main' into array-api-testing-docs
betatim May 12, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 18 additions & 3 deletions doc/modules/array_api.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -202,17 +202,32 @@ it supports the Array API. This will enable dedicated checks as part of the
common tests to verify that the estimators' results are the same when using
vanilla NumPy and Array API inputs.

To run the full set of checks you need to install both
`PyTorch <https://pytorch.org/>`_ and `CuPy <https://cupy.dev/>`_ and have
To run these checks you need to install
`array-api-strict <https://data-apis.org/array-api-strict/>`_ in your
test environment. This allows you to run checks without having a
GPU. To run the full set of checks you also need to install
`PyTorch <https://pytorch.org/>`_, `CuPy <https://cupy.dev/>`_ and have
a GPU. Checks that can not be executed or have missing dependencies will be
automatically skipped. Therefore it's important to run the tests with the
`-v` flag to see which checks are skipped:

.. prompt:: bash $

pip install ... # selected libraries as needed
pip install array-api-strict # and other libraries as needed
pytest -k "array_api" -v

Running the scikit-learn tests against `array-api-strict` should help reveal
most code problems related to handling multiple device inputs via the use of
simulated non-CPU devices. This allows for fast iterative development and debugging of
array API related code.

However, to ensure full handling of PyTorch or CuPy inputs allocated on actual GPU
devices, it is necessary to run the tests against those libraries and hardware.
This can either be achieved by using
`Google Colab <https://gist.github.com/EdAbati/ff3bdc06bafeb92452b3740686cc8d7c>`_
or leveraging our CI infrastructure on pull requests (manually triggered by maintainers
for cost reasons).

.. _mps_support:

Note on MPS device support
Expand Down
Loading