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

[breaking] Bump Python requirement to 3.10. #10434

Merged
merged 8 commits into from
Jul 30, 2024
Merged

[breaking] Bump Python requirement to 3.10. #10434

merged 8 commits into from
Jul 30, 2024

Conversation

trivialfis
Copy link
Member

@trivialfis trivialfis commented Jun 17, 2024 •
edited
Loading

  • Bump the Python requirement.
  • Fix type hints.
  • Use loky to avoid deadlock.
  • Workaround cupy-numpy compatibility issue on Windows caused by the safe casting rule.
  • Simplify the repartitioning logic to avoid dask errors.

Close #10371

@trivialfis
Copy link
Member Author

all modin tests are broken due to updated numpy.

@trivialfis
Copy link
Member Author

cupy is running into error as well.

@trivialfis
Copy link
Member Author

dask-expr appears to be running into trouble with the latest pandas.

@trivialfis
Copy link
Member Author

  • Moved the make sparse function into data.py to fix pylint error, also the definition is more consistent.
  • Removed the deterministic partitioning function in dask test due to dask internal error. This might make the test flaky, we will see.

@trivialfis
Copy link
Member Author

Err, immediately fail after removing the partitioning function.

@trivialfis
Copy link
Member Author

Couldn't reproduce the hang, will try it on aws later.

@trivialfis
Copy link
Member Author

trivialfis commented Jul 26, 2024 •
edited
Loading

@hcho3 Could you please help take another look? I have updated the PR description.

@trivialfis trivialfis merged commit 827d0e8 into dmlc:master Jul 30, 2024
28 checks passed
@trivialfis trivialfis deleted the py-310 branch July 30, 2024 09:31
@jakirkham jakirkham mentioned this pull request Jul 31, 2024
Closed
11 tasks
hcho3 added a commit to hcho3/xgboost that referenced this pull request Feb 4, 2025
@hcho3
[CI] Apply part of dmlc#10434
986068d
hcho3 added a commit that referenced this pull request Feb 4, 2025
@hcho3 @trivialfis @razdoburdin
[CI] Various CI fixes (#11196)
fc32798 
* Fix #10752 (#10972)

* Fix #10752

* [CI] Replace Mambaforge -> Miniforge3

* Fix formatting

* Increase timeout for Multi-GPU approx tests. (#11111)

* Fix tests with the latest scikit-learn. (#11086)

* Fix tests with the latest scikit-learn.

* dask.

* Remove scikit-learn pin

---------

Co-authored-by: Philip Hyunsu Cho <[email protected]>

* Revert base class change

* SYCL. Fix CI failures (#10987)

* [CI] Use Python 3.10 + latest sklearn in Windows

* Increase timeout in C++ tests from 1 to 5 seconds. (#10756)

To avoid CI failures on FreeBSD.

* [CI] Pin sklearn to 1.5.2

* [CI] Apply part of #10434

---------

Co-authored-by: Jiaming Yuan <[email protected]>
Co-authored-by: Dmitry Razdoburdin <[email protected]>
@jameslamb
Copy link
Contributor

dask-expr appears to be running into trouble with the latest pandas.

Not sure what specifically you saw when working on this, but just to connect some threads... I was trying to run xgboost v2.0.3 tests today with newer versions of Dask and the exact codepaths in test_with_dask.py modified here exposed what I think are some bugs that still exist in dask / dask-expr.

Linking:

@jakirkham jakirkham mentioned this pull request Mar 27, 2025
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@hcho3 hcho3 hcho3 approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Update Python requirement to 3.10.
3 participants
@trivialfis @jameslamb @hcho3