Widen scikit-learn constraint to <2 - #103
Merged
Merged
Conversation
The <=1.4.2 pin came from a blanket version-pinning commit; the only sklearn usage is Ridge from sklearn.linear_model, which is stable across 1.x releases. Match the style of the numpy/pandas constraints. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01FT7d8d2E3FansicmSzJ1SW
3 tasks
yury-lysogorskiy
added a commit
that referenced
this pull request
Aug 27, 2026
Both CI jobs aborted during pytest collection with
pyace/activelearning.py:9: from maxvolpy.maxvol import maxvol
E ModuleNotFoundError: No module named 'maxvolpy'
Interrupted: 2 errors during collection
so no test ran at all. This is pre-existing and unrelated to any particular
change - run 31264271622 (17 Aug, PR #103) fails identically.
`maxvolpy` is installed by the custom InstallMaxVolPyLocalPackage(install)
command in setup.py, which shells out to `cd lib/maxvolpy; python setup.py
install`. The workflow installs with `pip install .`, which builds a wheel and
never invokes the legacy `install` command, so that hook never runs.
(`python setup.py install` is also gone in setuptools >= 80.)
Build it explicitly instead. `--no-build-isolation` is required because
lib/maxvolpy/setup.py imports Cython and numpy at module level while declaring
no pyproject.toml build-system, so an isolated build environment fails with
ModuleNotFoundError: No module named 'Cython'. Cython and scipy are installed
first: the extension cimports scipy/linalg/cython_blas.pxd. numpy is
deliberately not installed here, to avoid resolving past setup.py's
`numpy<=1.26.4` pin; it is already present from the python-ace install.
Verified locally in a clean venv: with build isolation the build fails on the
missing Cython, without scipy it fails cythonizing _maxvol.pyx, and with the
command used here it builds a wheel and `from maxvolpy.maxvol import maxvol`
succeeds.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017QbfG2UVKxkeuiz9GtDLo4
srmnitc
added a commit
to pmrv/python-ace
that referenced
this pull request
Sep 4, 2026
Bring in the CI fixes that landed after this branch was cut, so the pipeline can actually exercise the pkg_resources change. Both jobs on this PR (run 31264233541, 8 Aug) failed in "Test with python-ace with tensorpotential". The logs are not readable without admin rights, but the failure is pre-existing and not from this branch: the sibling PR ICAMS#103 (run 31264271622) failed in the same step on the same day off the same base, and 3bfa4df records that failure as pyace/activelearning.py:9: from maxvolpy.maxvol import maxvol E ModuleNotFoundError: No module named 'maxvolpy' Interrupted: 2 errors during collection i.e. no test ran at all. Master has since fixed it: * 3bfa4df installs lib/maxvolpy explicitly, because `pip install .` builds a wheel and never runs setup.py's InstallMaxVolPyLocalPackage hook, so the suite could not be collected at all. This branch's run predates that step existing in the workflow. * d63466f imports UnitCellFilter from ase.filters with a fallback to ase.constraints; that was the sole remaining failure once collection worked, and it only affected the 3.10 job. Merge is clean. Both sides touch bin/pacemaker in different hunks - the import block here, the cutoff deprecation from f366916 further down - and both survived intact. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Bring in the CI fixes that landed after this branch was cut, so the pipeline can actually exercise the widened scikit-learn constraint. Both jobs on this PR (run 31264271622, 8 Aug) failed in "Test with python-ace with tensorpotential" - the run 3bfa4df cites by number as failing identically to ICAMS#104, from pyace/activelearning.py:9: from maxvolpy.maxvol import maxvol E ModuleNotFoundError: No module named 'maxvolpy' Interrupted: 2 errors during collection i.e. collection aborted and no test ran, so the sklearn bound was never tested either way. Pre-existing and unrelated to this branch. Master has since fixed it in 3bfa4df (install lib/maxvolpy explicitly, since `pip install .` builds a wheel and never runs setup.py's InstallMaxVolPyLocalPackage hook) and d63466f (UnitCellFilter moved to ase.filters; 3.10 job only). Merge is clean - master has not touched setup.py since b143ac3. Verified the premise of this change while I was here. sklearn is used in exactly one place, src/pyace/linearacefit.py:332: Ridge(alpha=1e-5, fit_intercept=False, copy_X=False, random_state=42, solver="auto") plus .fit(), .predict() and .coef_. Ran that exact call against the versions the two jobs will now resolve - 1.6.1 for 3.9 and 1.7.2 for 3.10 (1.8+ requires Python >= 3.11, so it is out of reach here) - and against 1.9.0 as the top of the range, each alongside the numpy<=1.26.4 pin from the same install_requires. Identical coefficients, no warnings, no resolver backtracking: sklearn 1.9.0 asks for numpy>=1.24.1, which the pin satisfies. No `normalize` argument anywhere, which is the one Ridge parameter removed during 1.x (in 1.2). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description of Changes
The <=1.4.2 pin came from a blanket version-pinning commit; the only sklearn usage is Ridge from sklearn.linear_model, which is stable across 1.x releases. Match the style of the numpy/pandas constraints.
Checklist
License Agreement
By submitting this pull request, I agree that:
Thank you for your contribution!