Up to [cvs.NetBSD.org] / pkgsrc / math / py-scikit-learn / patches
Request diff between arbitrary revisions
Keyword substitution: kv
Default branch: MAIN
py-scikit-learn: updated to 1.9.1 1.9.1 Support for Array API - |Fix| Fixed classification metrics to accept pandas labels when array API dispatch is enabled and predictions are from a different array API namespace or device. By :user:`Olivier Grisel <ogrisel>`. Metadata routing Refer to the :ref:`Metadata Routing User Guide <metadata_routing>` for more details. - |Fix| Metadata is now correctly routed to the sub-estimator of meta-estimators such as :class:`feature_selection.SelectFromModel`, :class:`feature_selection.RFE`, :class:`feature_selection.RFECV`, :class:`feature_selection.SequentialFeatureSelector`, and :class:`impute.IterativeImputer` when they are used as an intermediate step of a :class:`pipeline.Pipeline`, i.e. when their composite ``fit_transform`` method is the target of the routing. Previously, metadata such as ``sample_weight`` was silently dropped in this case. By :user:`Adrin Jalali <adrinjalali>` and :user:`Kyle Schluns <kschluns>`. - |Fix| :class:`ensemble.BaggingClassifier` now correctly routes metadata via `predict` and `predict_proba` dynamically to the corresponding method on the sub-estimator. By :user:`Stefanie Senger <StefanieSenger>`. - |Fix| :class:`pipeline.Pipeline`'s :meth:`~pipeline.Pipeline.fit_transform` and :meth:`~pipeline.Pipeline.fit_predict` now correctly apply ``transform_input`` to metadata routed to intermediate steps, matching :meth:`~pipeline.Pipeline.fit`. By :user:`Stefanie Senger <StefanieSenger>`. - |Fix| :class:`compose.TransformedTargetRegressor` now correctly routes metadata such as ``sample_weight`` to its default regressor (:class:`~linear_model.LinearRegression`) when metadata routing is enabled. By :user:`Stefanie Senger <StefanieSenger>`. Callbacks - |Fix| Fixed a bug where fitting an estimator with a :class:`callback.ScoringMonitor` and a parallel backend using multi-processing could fail with "Could not pickle the task to send it to the workers". By :user:`Jérémie du Boisberranger <jeremiedbb>`. :mod:`sklearn.cluster` - |Fix| :class:`cluster.DBSCAN` and :func:`cluster.dbscan` no longer raise a spurious :class:`~sklearn.exceptions.EfficiencyWarning` when given a precomputed sparse distance matrix. By :user:`Vadim <Tweezky66>`. - |Fix| :class:`cluster.OPTICS` no longer raises a :class:`~sklearn.exceptions.EfficiencyWarning` when fitted on a precomputed sparse distance matrix. By :user:`Dea María Léon <deamarialeon>`. :mod:`sklearn.datasets` - |Fix| :func:`datasets.fetch_openml` now correctly recovers from a corrupted download by re-downloading it, instead of raising an error. By :user:`Roman Yurchak <rth>`. - |Fix| :func:`datasets.fetch_file` now raises a `ValueError` when `local_filename` is a path instead of a plain filename. By :user:`Dea María Léon <DeaMariaLeon>`. :mod:`sklearn.discriminant_analysis` - |Fix| :class:`discriminant_analysis.LinearDiscriminantAnalysis` and :class:`discriminant_analysis.QuadraticDiscriminantAnalysis` now return the same results when using `solver="svd"` and `solver="eigen"`. The normalization for `solver="svd"` is changed to `1/n_samples` instead of `1/(n_samples - n_classes)` and `1/(n_samples - 1)`, respectively, which is consistent with the maximum likelihood covariance estimator (same as the `solver="eigen"` default). By :user:`Daniel Herrera-Esposito <dherrera1911>`. :mod:`sklearn.metrics` - |Efficiency| Scorers created with :func:`~metrics.make_scorer` no longer deep-copy metadata kwargs on each call, reducing memory usage during model selection and cross-validation. By :user:`Stefanie Senger <StefanieSenger>`. :mod:`sklearn.neighbors` - |Fix| :meth:`neighbors.NearestNeighbors.radius_neighbors` now correctly handles an array-like ``radius`` when ``n_jobs > 1`` with the ``ball_tree`` and ``kd_tree`` algorithms. By :user:`Nicolas Salvy <nicolassalvy>`. :mod:`sklearn.preprocessing` - |Fix| Fixed a bug in :class:`preprocessing.QuantileTransformer` where fitting on sparse data with `ignore_implicit_zeros=True` and subsampling could compute degenerate quantiles for some features. By :user:`Arthur Lacote <cakedev0>`. :mod:`sklearn.svm` - |Fix| Fixed a memory leak when calling the `fit` method of :class:`svm.LinearSVR`. By :user:`Jake VanderPlas <jakevdp>`. - |Fix| Fixed an out-of-bound write in the ``fit`` method of :class:`svm.LinearSVC` memory when using ``multi_class="crammer_singer"``. By :user:`Jake VanderPlas <jakevdp>`. :mod:`sklearn.utils` - |Fix| The HTML representation of estimators now truncates the output feature list to 100 entries to improve rendering performance for estimators with many output features. By :user:`Dea María Léon <deamarialeon>`. - |Fix| Array API supporting functions and estimators that handle mixed input namespaces now make negative-stride NumPy arrays contiguous before transferring them to PyTorch through DLPack, avoiding a Python process abort in ``torch.from_dlpack``. By :user:`Yin Li <Kevin-Li-2025>`. - |Fix| The copy-to-clipboard icons in the HTML representation of estimators are now rendered as ``<button>`` elements with an ``aria-label`` and a visible focus outline, instead of non-interactive ``<i>`` elements, so they are reachable and usable with a keyboard and screen readers. By :user:`Dea María Léon <deamarialeon>`. - |Fix| ``set_output(transform="polars")`` no longer emits a ``DeprecationWarning`` with ``polars>=1.42.1`` by using ``how="horizontal_extend"`` instead of the deprecated ``how="horizontal"`` when concatenating polars dataframes. By :user:`Felix Hoehle <fhoehle>`.
py-scikit-learn: fix build on NetBSD