daniel.biebert
/
sklearn
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
sklearn/doc/whats_new/v1.1.rst

1410 lines
64 KiB
ReStructuredText
Raw Permalink Normal View History

first commit
2024-08-05 09:32:03 +02:00
.. include:: _contributors.rst
.. currentmodule:: sklearn
.. _release_notes_1_1:
===========
Version 1.1
===========
For a short description of the main highlights of the release, please refer to
:ref:`sphx_glr_auto_examples_release_highlights_plot_release_highlights_1_1_0.py`.
.. include:: changelog_legend.inc
.. _changes_1_1_3:
Version 1.1.3
=============
**October 2022**
This bugfix release only includes fixes for compatibility with the latest
SciPy release >= 1.9.2. Notable changes include:
- |Fix| Include `msvcp140.dll` in the scikit-learn wheels since it has been
removed in the latest SciPy wheels.
:pr:`24631` by :user:`Chiara Marmo <cmarmo>`.
- |Enhancement| Create wheels for Python 3.11.
:pr:`24446` by :user:`Chiara Marmo <cmarmo>`.
Other bug fixes will be available in the next 1.2 release, which will be
released in the coming weeks.
Note that support for 32-bit Python on Windows has been dropped in this release. This
is due to the fact that SciPy 1.9.2 also dropped the support for that platform.
Windows users are advised to install the 64-bit version of Python instead.
.. _changes_1_1_2:
Version 1.1.2
=============
**August 2022**
Changed models
--------------
The following estimators and functions, when fit with the same data and
parameters, may produce different models from the previous version. This often
occurs due to changes in the modelling logic (bug fixes or enhancements), or in
random sampling procedures.
- |Fix| :class:`manifold.TSNE` now throws a `ValueError` when fit with
`perplexity>=n_samples` to ensure mathematical correctness of the algorithm.
:pr:`10805` by :user:`Mathias Andersen <MrMathias>` and
:pr:`23471` by :user:`Meekail Zain <micky774>`.
Changelog
---------
- |Fix| A default HTML representation is shown for meta-estimators with invalid
parameters. :pr:`24015` by `Thomas Fan`_.
- |Fix| Add support for F-contiguous arrays for estimators and functions whose back-end
have been changed in 1.1.
:pr:`23990` by :user:`Julien Jerphanion <jjerphan>`.
- |Fix| Wheels are now available for MacOS 10.9 and greater. :pr:`23833` by
`Thomas Fan`_.
:mod:`sklearn.base`
...................
- |Fix| The `get_params` method of the :class:`base.BaseEstimator` class now supports
estimators with `type`-type params that have the `get_params` method.
:pr:`24017` by :user:`Henry Sorsky <hsorsky>`.
:mod:`sklearn.cluster`
......................
- |Fix| Fixed a bug in :class:`cluster.Birch` that could trigger an error when splitting
a node if there are duplicates in the dataset.
:pr:`23395` by :user:`Jérémie du Boisberranger <jeremiedbb>`.
:mod:`sklearn.feature_selection`
................................
- |Fix| :class:`feature_selection.SelectFromModel` defaults to selection
threshold 1e-5 when the estimator is either :class:`linear_model.ElasticNet`
or :class:`linear_model.ElasticNetCV` with `l1_ratio` equals 1 or
:class:`linear_model.LassoCV`.
:pr:`23636` by :user:`Hao Chun Chang <haochunchang>`.
:mod:`sklearn.impute`
.....................
- |Fix| :class:`impute.SimpleImputer` uses the dtype seen in `fit` for
`transform` when the dtype is object. :pr:`22063` by `Thomas Fan`_.
:mod:`sklearn.linear_model`
...........................
- |Fix| Use dtype-aware tolerances for the validation of gram matrices (passed by users
or precomputed). :pr:`22059` by :user:`Malte S. Kurz <MalteKurz>`.
- |Fix| Fixed an error in :class:`linear_model.LogisticRegression` with
`solver="newton-cg"`, `fit_intercept=True`, and a single feature. :pr:`23608`
by `Tom Dupre la Tour`_.
:mod:`sklearn.manifold`
.......................
- |Fix| :class:`manifold.TSNE` now throws a `ValueError` when fit with
`perplexity>=n_samples` to ensure mathematical correctness of the algorithm.
:pr:`10805` by :user:`Mathias Andersen <MrMathias>` and
:pr:`23471` by :user:`Meekail Zain <micky774>`.
:mod:`sklearn.metrics`
......................
- |Fix| Fixed error message of :class:`metrics.coverage_error` for 1D array input.
:pr:`23548` by :user:`Hao Chun Chang <haochunchang>`.
:mod:`sklearn.preprocessing`
............................
- |Fix| :meth:`preprocessing.OrdinalEncoder.inverse_transform` correctly handles
use cases where `unknown_value` or `encoded_missing_value` is `nan`. :pr:`24087`
by `Thomas Fan`_.
:mod:`sklearn.tree`
...................
- |Fix| Fixed invalid memory access bug during fit in
:class:`tree.DecisionTreeRegressor` and :class:`tree.DecisionTreeClassifier`.
:pr:`23273` by `Thomas Fan`_.
.. _changes_1_1_1:
Version 1.1.1
=============
**May 2022**
Changelog
---------
- |Enhancement| The error message is improved when importing
:class:`model_selection.HalvingGridSearchCV`,
:class:`model_selection.HalvingRandomSearchCV`, or
:class:`impute.IterativeImputer` without importing the experimental flag.
:pr:`23194` by `Thomas Fan`_.
- |Enhancement| Added an extension in doc/conf.py to automatically generate
the list of estimators that handle NaN values.
:pr:`23198` by :user:`Lise Kleiber <lisekleiber>`, :user:`Zhehao Liu <MaxwellLZH>`
and :user:`Chiara Marmo <cmarmo>`.
:mod:`sklearn.datasets`
.......................
- |Fix| Avoid timeouts in :func:`datasets.fetch_openml` by not passing a
`timeout` argument, :pr:`23358` by :user:`Loïc Estève <lesteve>`.
:mod:`sklearn.decomposition`
............................
- |Fix| Avoid spurious warning in :class:`decomposition.IncrementalPCA` when
`n_samples == n_components`. :pr:`23264` by :user:`Lucy Liu <lucyleeow>`.
:mod:`sklearn.feature_selection`
................................
- |Fix| The `partial_fit` method of :class:`feature_selection.SelectFromModel`
now conducts validation for `max_features` and `feature_names_in` parameters.
:pr:`23299` by :user:`Long Bao <lorentzbao>`.
:mod:`sklearn.metrics`
......................
- |Fix| Fixes :func:`metrics.precision_recall_curve` to compute precision-recall at 100%
recall. The Precision-Recall curve now displays the last point corresponding to a
classifier that always predicts the positive class: recall=100% and
precision=class balance.
:pr:`23214` by :user:`Stéphane Collot <stephanecollot>` and :user:`Max Baak <mbaak>`.
:mod:`sklearn.preprocessing`
............................
- |Fix| :class:`preprocessing.PolynomialFeatures` with ``degree`` equal to 0
will raise error when ``include_bias`` is set to False, and outputs a single
constant array when ``include_bias`` is set to True.
:pr:`23370` by :user:`Zhehao Liu <MaxwellLZH>`.
:mod:`sklearn.tree`
...................
- |Fix| Fixes performance regression with low cardinality features for
:class:`tree.DecisionTreeClassifier`,
:class:`tree.DecisionTreeRegressor`,
:class:`ensemble.RandomForestClassifier`,
:class:`ensemble.RandomForestRegressor`,
:class:`ensemble.GradientBoostingClassifier`, and
:class:`ensemble.GradientBoostingRegressor`.
:pr:`23410` by :user:`Loïc Estève <lesteve>`.
:mod:`sklearn.utils`
....................
- |Fix| :func:`utils.class_weight.compute_sample_weight` now works with sparse `y`.
:pr:`23115` by :user:`kernc <kernc>`.
.. _changes_1_1:
Version 1.1.0
=============
**May 2022**
Minimal dependencies
--------------------
Version 1.1.0 of scikit-learn requires python 3.8+, numpy 1.17.3+ and
scipy 1.3.2+. Optional minimal dependency is matplotlib 3.1.2+.
Changed models
--------------
The following estimators and functions, when fit with the same data and
parameters, may produce different models from the previous version. This often
occurs due to changes in the modelling logic (bug fixes or enhancements), or in
random sampling procedures.
- |Efficiency| :class:`cluster.KMeans` now defaults to ``algorithm="lloyd"``
instead of ``algorithm="auto"``, which was equivalent to
``algorithm="elkan"``. Lloyd's algorithm and Elkan's algorithm converge to the
same solution, up to numerical rounding errors, but in general Lloyd's
algorithm uses much less memory, and it is often faster.
- |Efficiency| Fitting :class:`tree.DecisionTreeClassifier`,
:class:`tree.DecisionTreeRegressor`,
:class:`ensemble.RandomForestClassifier`,
:class:`ensemble.RandomForestRegressor`,
:class:`ensemble.GradientBoostingClassifier`, and
:class:`ensemble.GradientBoostingRegressor` is on average 15% faster than in
previous versions thanks to a new sort algorithm to find the best split.
Models might be different because of a different handling of splits
with tied criterion values: both the old and the new sorting algorithm
are unstable sorting algorithms. :pr:`22868` by `Thomas Fan`_.
- |Fix| The eigenvectors initialization for :class:`cluster.SpectralClustering`
and :class:`manifold.SpectralEmbedding` now samples from a Gaussian when
using the `'amg'` or `'lobpcg'` solver. This change improves numerical
stability of the solver, but may result in a different model.
- |Fix| :func:`feature_selection.f_regression` and
:func:`feature_selection.r_regression` will now returned finite score by
default instead of `np.nan` and `np.inf` for some corner case. You can use
`force_finite=False` if you really want to get non-finite values and keep
the old behavior.
- |Fix| Panda's DataFrames with all non-string columns such as a MultiIndex no
longer warns when passed into an Estimator. Estimators will continue to
ignore the column names in DataFrames with non-string columns. For
`feature_names_in_` to be defined, columns must be all strings. :pr:`22410` by
`Thomas Fan`_.
- |Fix| :class:`preprocessing.KBinsDiscretizer` changed handling of bin edges
slightly, which might result in a different encoding with the same data.
- |Fix| :func:`calibration.calibration_curve` changed handling of bin
edges slightly, which might result in a different output curve given the same
data.
- |Fix| :class:`discriminant_analysis.LinearDiscriminantAnalysis` now uses
the correct variance-scaling coefficient which may result in different model
behavior.
- |Fix| :meth:`feature_selection.SelectFromModel.fit` and
:meth:`feature_selection.SelectFromModel.partial_fit` can now be called with
`prefit=True`. `estimators_` will be a deep copy of `estimator` when
`prefit=True`. :pr:`23271` by :user:`Guillaume Lemaitre <glemaitre>`.
Changelog
---------
..
Entries should be grouped by module (in alphabetic order) and prefixed with
one of the labels: |MajorFeature|, |Feature|, |Efficiency|, |Enhancement|,
|Fix| or |API| (see whats_new.rst for descriptions).
Entries should be ordered by those labels (e.g. |Fix| after |Efficiency|).
Changes not specific to a module should be listed under *Multiple Modules*
or *Miscellaneous*.
Entries should end with:
:pr:`123456` by :user:`Joe Bloggs <joeongithub>`.
where 123456 is the *pull request* number, not the issue number.
- |Efficiency| Low-level routines for reductions on pairwise distances
for dense float64 datasets have been refactored. The following functions
and estimators now benefit from improved performances in terms of hardware
scalability and speed-ups:
- :func:`sklearn.metrics.pairwise_distances_argmin`
- :func:`sklearn.metrics.pairwise_distances_argmin_min`
- :class:`sklearn.cluster.AffinityPropagation`
- :class:`sklearn.cluster.Birch`
- :class:`sklearn.cluster.MeanShift`
- :class:`sklearn.cluster.OPTICS`
- :class:`sklearn.cluster.SpectralClustering`
- :func:`sklearn.feature_selection.mutual_info_regression`
- :class:`sklearn.neighbors.KNeighborsClassifier`
- :class:`sklearn.neighbors.KNeighborsRegressor`
- :class:`sklearn.neighbors.RadiusNeighborsClassifier`
- :class:`sklearn.neighbors.RadiusNeighborsRegressor`
- :class:`sklearn.neighbors.LocalOutlierFactor`
- :class:`sklearn.neighbors.NearestNeighbors`
- :class:`sklearn.manifold.Isomap`
- :class:`sklearn.manifold.LocallyLinearEmbedding`
- :class:`sklearn.manifold.TSNE`
- :func:`sklearn.manifold.trustworthiness`
- :class:`sklearn.semi_supervised.LabelPropagation`
- :class:`sklearn.semi_supervised.LabelSpreading`
For instance :class:`sklearn.neighbors.NearestNeighbors.kneighbors` and
:class:`sklearn.neighbors.NearestNeighbors.radius_neighbors`
can respectively be up to ×20 and ×5 faster than previously on a laptop.
Moreover, implementations of those two algorithms are now suitable
for machine with many cores, making them usable for datasets consisting
of millions of samples.
:pr:`21987`, :pr:`22064`, :pr:`22065`, :pr:`22288` and :pr:`22320`
by :user:`Julien Jerphanion <jjerphan>`.
- |Enhancement| All scikit-learn models now generate a more informative
error message when some input contains unexpected `NaN` or infinite values.
In particular the message contains the input name ("X", "y" or
"sample_weight") and if an unexpected `NaN` value is found in `X`, the error
message suggests potential solutions.
:pr:`21219` by :user:`Olivier Grisel <ogrisel>`.
- |Enhancement| All scikit-learn models now generate a more informative
error message when setting invalid hyper-parameters with `set_params`.
:pr:`21542` by :user:`Olivier Grisel <ogrisel>`.
- |Enhancement| Removes random unique identifiers in the HTML representation.
With this change, jupyter notebooks are reproducible as long as the cells are
run in the same order. :pr:`23098` by `Thomas Fan`_.
- |Fix| Estimators with `non_deterministic` tag set to `True` will skip both
`check_methods_sample_order_invariance` and `check_methods_subset_invariance` tests.
:pr:`22318` by :user:`Zhehao Liu <MaxwellLZH>`.
- |API| The option for using the log loss, aka binomial or multinomial deviance, via
the `loss` parameters was made more consistent. The preferred way is by
setting the value to `"log_loss"`. Old option names are still valid and
produce the same models, but are deprecated and will be removed in version
1.3.
- For :class:`ensemble.GradientBoostingClassifier`, the `loss` parameter name
"deviance" is deprecated in favor of the new name "log_loss", which is now the
default.
:pr:`23036` by :user:`Christian Lorentzen <lorentzenchr>`.
- For :class:`ensemble.HistGradientBoostingClassifier`, the `loss` parameter names
"auto", "binary_crossentropy" and "categorical_crossentropy" are deprecated in
favor of the new name "log_loss", which is now the default.
:pr:`23040` by :user:`Christian Lorentzen <lorentzenchr>`.
- For :class:`linear_model.SGDClassifier`, the `loss` parameter name
"log" is deprecated in favor of the new name "log_loss".
:pr:`23046` by :user:`Christian Lorentzen <lorentzenchr>`.
- |API| Rich html representation of estimators is now enabled by default in Jupyter
notebooks. It can be deactivated by setting `display='text'` in
:func:`sklearn.set_config`.
:pr:`22856` by :user:`Jérémie du Boisberranger <jeremiedbb>`.
:mod:`sklearn.calibration`
..........................
- |Enhancement| :func:`calibration.calibration_curve` accepts a parameter
`pos_label` to specify the positive class label.
:pr:`21032` by :user:`Guillaume Lemaitre <glemaitre>`.
- |Enhancement| :meth:`calibration.CalibratedClassifierCV.fit` now supports passing
`fit_params`, which are routed to the `base_estimator`.
:pr:`18170` by :user:`Benjamin Bossan <BenjaminBossan>`.
- |Enhancement| :class:`calibration.CalibrationDisplay` accepts a parameter `pos_label`
to add this information to the plot.
:pr:`21038` by :user:`Guillaume Lemaitre <glemaitre>`.
- |Fix| :func:`calibration.calibration_curve` handles bin edges more consistently now.
:pr:`14975` by `Andreas Müller`_ and :pr:`22526` by :user:`Meekail Zain <micky774>`.
- |API| :func:`calibration.calibration_curve`'s `normalize` parameter is
now deprecated and will be removed in version 1.3. It is recommended that
a proper probability (i.e. a classifier's :term:`predict_proba` positive
class) is used for `y_prob`.
:pr:`23095` by :user:`Jordan Silke <jsilke>`.
:mod:`sklearn.cluster`
......................
- |MajorFeature| :class:`cluster.BisectingKMeans` introducing Bisecting K-Means algorithm
:pr:`20031` by :user:`Michal Krawczyk <michalkrawczyk>`,
:user:`Tom Dupre la Tour <TomDLT>`
and :user:`Jérémie du Boisberranger <jeremiedbb>`.
- |Enhancement| :class:`cluster.SpectralClustering` and
:func:`cluster.spectral_clustering` now include the new `'cluster_qr'` method that
clusters samples in the embedding space as an alternative to the existing `'kmeans'`
and `'discrete'` methods. See :func:`cluster.spectral_clustering` for more details.
:pr:`21148` by :user:`Andrew Knyazev <lobpcg>`.
- |Enhancement| Adds :term:`get_feature_names_out` to :class:`cluster.Birch`,
:class:`cluster.FeatureAgglomeration`, :class:`cluster.KMeans`,
:class:`cluster.MiniBatchKMeans`. :pr:`22255` by `Thomas Fan`_.
- |Enhancement| :class:`cluster.SpectralClustering` now raises consistent
error messages when passed invalid values for `n_clusters`, `n_init`,
`gamma`, `n_neighbors`, `eigen_tol` or `degree`.
:pr:`21881` by :user:`Hugo Vassard <hvassard>`.
- |Enhancement| :class:`cluster.AffinityPropagation` now returns cluster
centers and labels if they exist, even if the model has not fully converged.
When returning these potentially-degenerate cluster centers and labels, a new
warning message is shown. If no cluster centers were constructed,
then the cluster centers remain an empty list with labels set to
`-1` and the original warning message is shown.
:pr:`22217` by :user:`Meekail Zain <micky774>`.
- |Efficiency| In :class:`cluster.KMeans`, the default ``algorithm`` is now
``"lloyd"`` which is the full classical EM-style algorithm. Both ``"auto"``
and ``"full"`` are deprecated and will be removed in version 1.3. They are
now aliases for ``"lloyd"``. The previous default was ``"auto"``, which relied
on Elkan's algorithm. Lloyd's algorithm uses less memory than Elkan's, it
is faster on many datasets, and its results are identical, hence the change.
:pr:`21735` by :user:`Aurélien Geron <ageron>`.
- |Fix| :class:`cluster.KMeans`'s `init` parameter now properly supports
array-like input and NumPy string scalars. :pr:`22154` by `Thomas Fan`_.
:mod:`sklearn.compose`
......................
- |Fix| :class:`compose.ColumnTransformer` now removes validation errors from
`__init__` and `set_params` methods.
:pr:`22537` by :user:`iofall <iofall>` and :user:`Arisa Y. <arisayosh>`.
- |Fix| :term:`get_feature_names_out` functionality in
:class:`compose.ColumnTransformer` was broken when columns were specified
using `slice`. This is fixed in :pr:`22775` and :pr:`22913` by
:user:`randomgeek78 <randomgeek78>`.
:mod:`sklearn.covariance`
.........................
- |Fix| :class:`covariance.GraphicalLassoCV` now accepts NumPy array for the
parameter `alphas`.
:pr:`22493` by :user:`Guillaume Lemaitre <glemaitre>`.
:mod:`sklearn.cross_decomposition`
..................................
- |Enhancement| the `inverse_transform` method of
:class:`cross_decomposition.PLSRegression`, :class:`cross_decomposition.PLSCanonical`
and :class:`cross_decomposition.CCA` now allows reconstruction of a `X` target when
a `Y` parameter is given. :pr:`19680` by
:user:`Robin Thibaut <robinthibaut>`.
- |Enhancement| Adds :term:`get_feature_names_out` to all transformers in the
:mod:`~sklearn.cross_decomposition` module:
:class:`cross_decomposition.CCA`,
:class:`cross_decomposition.PLSSVD`,
:class:`cross_decomposition.PLSRegression`,
and :class:`cross_decomposition.PLSCanonical`. :pr:`22119` by `Thomas Fan`_.
- |Fix| The shape of the :term:`coef_` attribute of :class:`cross_decomposition.CCA`,
:class:`cross_decomposition.PLSCanonical` and
:class:`cross_decomposition.PLSRegression` will change in version 1.3, from
`(n_features, n_targets)` to `(n_targets, n_features)`, to be consistent
with other linear models and to make it work with interface expecting a
specific shape for `coef_` (e.g. :class:`feature_selection.RFE`).
:pr:`22016` by :user:`Guillaume Lemaitre <glemaitre>`.
- |API| add the fitted attribute `intercept_` to
:class:`cross_decomposition.PLSCanonical`,
:class:`cross_decomposition.PLSRegression`, and
:class:`cross_decomposition.CCA`. The method `predict` is indeed equivalent to
`Y = X @ coef_ + intercept_`.
:pr:`22015` by :user:`Guillaume Lemaitre <glemaitre>`.
:mod:`sklearn.datasets`
.......................
- |Feature| :func:`datasets.load_files` now accepts a ignore list and
an allow list based on file extensions.
:pr:`19747` by :user:`Tony Attalla <tonyattalla>` and :pr:`22498` by
:user:`Meekail Zain <micky774>`.
- |Enhancement| :func:`datasets.make_swiss_roll` now supports the optional argument
hole; when set to True, it returns the swiss-hole dataset. :pr:`21482` by
:user:`Sebastian Pujalte <pujaltes>`.
- |Enhancement| :func:`datasets.make_blobs` no longer copies data during the generation
process, therefore uses less memory.
:pr:`22412` by :user:`Zhehao Liu <MaxwellLZH>`.
- |Enhancement| :func:`datasets.load_diabetes` now accepts the parameter
``scaled``, to allow loading unscaled data. The scaled version of this
dataset is now computed from the unscaled data, and can produce slightly
different results that in previous version (within a 1e-4 absolute
tolerance).
:pr:`16605` by :user:`Mandy Gu <happilyeverafter95>`.
- |Enhancement| :func:`datasets.fetch_openml` now has two optional arguments
`n_retries` and `delay`. By default, :func:`datasets.fetch_openml` will retry
3 times in case of a network failure with a delay between each try.
:pr:`21901` by :user:`Rileran <rileran>`.
- |Fix| :func:`datasets.fetch_covtype` is now concurrent-safe: data is downloaded
to a temporary directory before being moved to the data directory.
:pr:`23113` by :user:`Ilion Beyst <iasoon>`.
- |API| :func:`datasets.make_sparse_coded_signal` now accepts a parameter
`data_transposed` to explicitly specify the shape of matrix `X`. The default
behavior `True` is to return a transposed matrix `X` corresponding to a
`(n_features, n_samples)` shape. The default value will change to `False` in
version 1.3. :pr:`21425` by :user:`Gabriel Stefanini Vicente <g4brielvs>`.
:mod:`sklearn.decomposition`
............................
- |MajorFeature| Added a new estimator :class:`decomposition.MiniBatchNMF`. It is a
faster but less accurate version of non-negative matrix factorization, better suited
for large datasets. :pr:`16948` by :user:`Chiara Marmo <cmarmo>`,
:user:`Patricio Cerda <pcerda>` and :user:`Jérémie du Boisberranger <jeremiedbb>`.
- |Enhancement| :func:`decomposition.dict_learning`,
:func:`decomposition.dict_learning_online`
and :func:`decomposition.sparse_encode` preserve dtype for `numpy.float32`.
:class:`decomposition.DictionaryLearning`,
:class:`decomposition.MiniBatchDictionaryLearning`
and :class:`decomposition.SparseCoder` preserve dtype for `numpy.float32`.
:pr:`22002` by :user:`Takeshi Oura <takoika>`.
- |Enhancement| :class:`decomposition.PCA` exposes a parameter `n_oversamples` to tune
:func:`utils.extmath.randomized_svd` and get accurate results when the number of
features is large.
:pr:`21109` by :user:`Smile <x-shadow-man>`.
- |Enhancement| The :class:`decomposition.MiniBatchDictionaryLearning` and
:func:`decomposition.dict_learning_online` have been refactored and now have a
stopping criterion based on a small change of the dictionary or objective function,
controlled by the new `max_iter`, `tol` and `max_no_improvement` parameters. In
addition, some of their parameters and attributes are deprecated.
- the `n_iter` parameter of both is deprecated. Use `max_iter` instead.
- the `iter_offset`, `return_inner_stats`, `inner_stats` and `return_n_iter`
parameters of :func:`decomposition.dict_learning_online` serve internal purpose
and are deprecated.
- the `inner_stats_`, `iter_offset_` and `random_state_` attributes of
:class:`decomposition.MiniBatchDictionaryLearning` serve internal purpose and are
deprecated.
- the default value of the `batch_size` parameter of both will change from 3 to 256
in version 1.3.
:pr:`18975` by :user:`Jérémie du Boisberranger <jeremiedbb>`.
- |Enhancement| :class:`decomposition.SparsePCA` and :class:`decomposition.MiniBatchSparsePCA`
preserve dtype for `numpy.float32`.
:pr:`22111` by :user:`Takeshi Oura <takoika>`.
- |Enhancement| :class:`decomposition.TruncatedSVD` now allows
`n_components == n_features`, if `algorithm='randomized'`.
:pr:`22181` by :user:`Zach Deane-Mayer <zachmayer>`.
- |Enhancement| Adds :term:`get_feature_names_out` to all transformers in the
:mod:`~sklearn.decomposition` module:
:class:`decomposition.DictionaryLearning`,
:class:`decomposition.FactorAnalysis`,
:class:`decomposition.FastICA`,
:class:`decomposition.IncrementalPCA`,
:class:`decomposition.KernelPCA`,
:class:`decomposition.LatentDirichletAllocation`,
:class:`decomposition.MiniBatchDictionaryLearning`,
:class:`decomposition.MiniBatchSparsePCA`,
:class:`decomposition.NMF`,
:class:`decomposition.PCA`,
:class:`decomposition.SparsePCA`,
and :class:`decomposition.TruncatedSVD`. :pr:`21334` by
`Thomas Fan`_.
- |Enhancement| :class:`decomposition.TruncatedSVD` exposes the parameter
`n_oversamples` and `power_iteration_normalizer` to tune
:func:`utils.extmath.randomized_svd` and get accurate results when the number
of features is large, the rank of the matrix is high, or other features of
the matrix make low rank approximation difficult.
:pr:`21705` by :user:`Jay S. Stanley III <stanleyjs>`.
- |Enhancement| :class:`decomposition.PCA` exposes the parameter
`power_iteration_normalizer` to tune :func:`utils.extmath.randomized_svd` and
get more accurate results when low rank approximation is difficult.
:pr:`21705` by :user:`Jay S. Stanley III <stanleyjs>`.
- |Fix| :class:`decomposition.FastICA` now validates input parameters in `fit`
instead of `__init__`.
:pr:`21432` by :user:`Hannah Bohle <hhnnhh>` and
:user:`Maren Westermann <marenwestermann>`.
- |Fix| :class:`decomposition.FastICA` now accepts `np.float32` data without
silent upcasting. The dtype is preserved by `fit` and `fit_transform` and the
main fitted attributes use a dtype of the same precision as the training
data. :pr:`22806` by :user:`Jihane Bennis <JihaneBennis>` and
:user:`Olivier Grisel <ogrisel>`.
- |Fix| :class:`decomposition.FactorAnalysis` now validates input parameters
in `fit` instead of `__init__`.
:pr:`21713` by :user:`Haya <HayaAlmutairi>` and :user:`Krum Arnaudov <krumeto>`.
- |Fix| :class:`decomposition.KernelPCA` now validates input parameters in
`fit` instead of `__init__`.
:pr:`21567` by :user:`Maggie Chege <MaggieChege>`.
- |Fix| :class:`decomposition.PCA` and :class:`decomposition.IncrementalPCA`
more safely calculate precision using the inverse of the covariance matrix
if `self.noise_variance_` is zero.
:pr:`22300` by :user:`Meekail Zain <micky774>` and :pr:`15948` by :user:`sysuresh`.
- |Fix| Greatly reduced peak memory usage in :class:`decomposition.PCA` when
calling `fit` or `fit_transform`.
:pr:`22553` by :user:`Meekail Zain <micky774>`.
- |API| :func:`decomposition.FastICA` now supports unit variance for whitening.
The default value of its `whiten` argument will change from `True`
(which behaves like `'arbitrary-variance'`) to `'unit-variance'` in version 1.3.
:pr:`19490` by :user:`Facundo Ferrin <fferrin>` and
:user:`Julien Jerphanion <jjerphan>`.
:mod:`sklearn.discriminant_analysis`
....................................
- |Enhancement| Adds :term:`get_feature_names_out` to
:class:`discriminant_analysis.LinearDiscriminantAnalysis`. :pr:`22120` by
`Thomas Fan`_.
- |Fix| :class:`discriminant_analysis.LinearDiscriminantAnalysis` now uses
the correct variance-scaling coefficient which may result in different model
behavior. :pr:`15984` by :user:`Okon Samuel <OkonSamuel>` and :pr:`22696` by
:user:`Meekail Zain <micky774>`.
:mod:`sklearn.dummy`
....................
- |Fix| :class:`dummy.DummyRegressor` no longer overrides the `constant`
parameter during `fit`. :pr:`22486` by `Thomas Fan`_.
:mod:`sklearn.ensemble`
.......................
- |MajorFeature| Added additional option `loss="quantile"` to
:class:`ensemble.HistGradientBoostingRegressor` for modelling quantiles.
The quantile level can be specified with the new parameter `quantile`.
:pr:`21800` and :pr:`20567` by :user:`Christian Lorentzen <lorentzenchr>`.
- |Efficiency| `fit` of :class:`ensemble.GradientBoostingClassifier`
and :class:`ensemble.GradientBoostingRegressor` now calls :func:`utils.check_array`
with parameter `force_all_finite=False` for non initial warm-start runs as it has
already been checked before.
:pr:`22159` by :user:`Geoffrey Paris <Geoffrey-Paris>`.
- |Enhancement| :class:`ensemble.HistGradientBoostingClassifier` is faster,
for binary and in particular for multiclass problems thanks to the new private loss
function module.
:pr:`20811`, :pr:`20567` and :pr:`21814` by
:user:`Christian Lorentzen <lorentzenchr>`.
- |Enhancement| Adds support to use pre-fit models with `cv="prefit"`
in :class:`ensemble.StackingClassifier` and :class:`ensemble.StackingRegressor`.
:pr:`16748` by :user:`Siqi He <siqi-he>` and :pr:`22215` by
:user:`Meekail Zain <micky774>`.
- |Enhancement| :class:`ensemble.RandomForestClassifier` and
:class:`ensemble.ExtraTreesClassifier` have the new `criterion="log_loss"`, which is
equivalent to `criterion="entropy"`.
:pr:`23047` by :user:`Christian Lorentzen <lorentzenchr>`.
- |Enhancement| Adds :term:`get_feature_names_out` to
:class:`ensemble.VotingClassifier`, :class:`ensemble.VotingRegressor`,
:class:`ensemble.StackingClassifier`, and
:class:`ensemble.StackingRegressor`. :pr:`22695` and :pr:`22697` by `Thomas Fan`_.
- |Enhancement| :class:`ensemble.RandomTreesEmbedding` now has an informative
:term:`get_feature_names_out` function that includes both tree index and leaf index in
the output feature names.
:pr:`21762` by :user:`Zhehao Liu <MaxwellLZH>` and `Thomas Fan`_.
- |Efficiency| Fitting a :class:`ensemble.RandomForestClassifier`,
:class:`ensemble.RandomForestRegressor`, :class:`ensemble.ExtraTreesClassifier`,
:class:`ensemble.ExtraTreesRegressor`, and :class:`ensemble.RandomTreesEmbedding`
is now faster in a multiprocessing setting, especially for subsequent fits with
`warm_start` enabled.
:pr:`22106` by :user:`Pieter Gijsbers <PGijsbers>`.
- |Fix| Change the parameter `validation_fraction` in
:class:`ensemble.GradientBoostingClassifier` and
:class:`ensemble.GradientBoostingRegressor` so that an error is raised if anything
other than a float is passed in as an argument.
:pr:`21632` by :user:`Genesis Valencia <genvalen>`.
- |Fix| Removed a potential source of CPU oversubscription in
:class:`ensemble.HistGradientBoostingClassifier` and
:class:`ensemble.HistGradientBoostingRegressor` when CPU resource usage is limited,
for instance using cgroups quota in a docker container. :pr:`22566` by
:user:`Jérémie du Boisberranger <jeremiedbb>`.
- |Fix| :class:`ensemble.HistGradientBoostingClassifier` and
:class:`ensemble.HistGradientBoostingRegressor` no longer warns when
fitting on a pandas DataFrame with a non-default `scoring` parameter and
early_stopping enabled. :pr:`22908` by `Thomas Fan`_.
- |Fix| Fixes HTML repr for :class:`ensemble.StackingClassifier` and
:class:`ensemble.StackingRegressor`. :pr:`23097` by `Thomas Fan`_.
- |API| The attribute `loss_` of :class:`ensemble.GradientBoostingClassifier` and
:class:`ensemble.GradientBoostingRegressor` has been deprecated and will be removed
in version 1.3.
:pr:`23079` by :user:`Christian Lorentzen <lorentzenchr>`.
- |API| Changed the default of `max_features` to 1.0 for
:class:`ensemble.RandomForestRegressor` and to `"sqrt"` for
:class:`ensemble.RandomForestClassifier`. Note that these give the same fit
results as before, but are much easier to understand. The old default value
`"auto"` has been deprecated and will be removed in version 1.3. The same
changes are also applied for :class:`ensemble.ExtraTreesRegressor` and
:class:`ensemble.ExtraTreesClassifier`.
:pr:`20803` by :user:`Brian Sun <bsun94>`.
- |Efficiency| Improve runtime performance of :class:`ensemble.IsolationForest`
by skipping repetitive input checks. :pr:`23149` by :user:`Zhehao Liu <MaxwellLZH>`.
:mod:`sklearn.feature_extraction`
.................................
- |Feature| :class:`feature_extraction.FeatureHasher` now supports PyPy.
:pr:`23023` by `Thomas Fan`_.
- |Fix| :class:`feature_extraction.FeatureHasher` now validates input parameters
in `transform` instead of `__init__`. :pr:`21573` by
:user:`Hannah Bohle <hhnnhh>` and :user:`Maren Westermann <marenwestermann>`.
- |Fix| :class:`feature_extraction.text.TfidfVectorizer` now does not create
a :class:`feature_extraction.text.TfidfTransformer` at `__init__` as required
by our API.
:pr:`21832` by :user:`Guillaume Lemaitre <glemaitre>`.
:mod:`sklearn.feature_selection`
................................
- |Feature| Added auto mode to :class:`feature_selection.SequentialFeatureSelector`.
If the argument `n_features_to_select` is `'auto'`, select features until the score
improvement does not exceed the argument `tol`. The default value of
`n_features_to_select` changed from `None` to `'warn'` in 1.1 and will become
`'auto'` in 1.3. `None` and `'warn'` will be removed in 1.3. :pr:`20145` by
:user:`murata-yu <murata-yu>`.
- |Feature| Added the ability to pass callables to the `max_features` parameter
of :class:`feature_selection.SelectFromModel`. Also introduced new attribute
`max_features_` which is inferred from `max_features` and the data during
`fit`. If `max_features` is an integer, then `max_features_ = max_features`.
If `max_features` is a callable, then `max_features_ = max_features(X)`.
:pr:`22356` by :user:`Meekail Zain <micky774>`.
- |Enhancement| :class:`feature_selection.GenericUnivariateSelect` preserves
float32 dtype. :pr:`18482` by :user:`Thierry Gameiro <titigmr>`
and :user:`Daniel Kharsa <aflatoune>` and :pr:`22370` by
:user:`Meekail Zain <micky774>`.
- |Enhancement| Add a parameter `force_finite` to
:func:`feature_selection.f_regression` and
:func:`feature_selection.r_regression`. This parameter allows to force the
output to be finite in the case where a feature or a the target is constant
or that the feature and target are perfectly correlated (only for the
F-statistic).
:pr:`17819` by :user:`Juan Carlos Alfaro Jiménez <alfaro96>`.
- |Efficiency| Improve runtime performance of :func:`feature_selection.chi2`
with boolean arrays. :pr:`22235` by `Thomas Fan`_.
- |Efficiency| Reduced memory usage of :func:`feature_selection.chi2`.
:pr:`21837` by :user:`Louis Wagner <lrwagner>`.
:mod:`sklearn.gaussian_process`
...............................
- |Fix| `predict` and `sample_y` methods of
:class:`gaussian_process.GaussianProcessRegressor` now return
arrays of the correct shape in single-target and multi-target cases, and for
both `normalize_y=False` and `normalize_y=True`.
:pr:`22199` by :user:`Guillaume Lemaitre <glemaitre>`,
:user:`Aidar Shakerimoff <AidarShakerimoff>` and
:user:`Tenavi Nakamura-Zimmerer <Tenavi>`.
- |Fix| :class:`gaussian_process.GaussianProcessClassifier` raises
a more informative error if `CompoundKernel` is passed via `kernel`.
:pr:`22223` by :user:`MarcoM <marcozzxx810>`.
:mod:`sklearn.impute`
.....................
- |Enhancement| :class:`impute.SimpleImputer` now warns with feature names when features
which are skipped due to the lack of any observed values in the training set.
:pr:`21617` by :user:`Christian Ritter <chritter>`.
- |Enhancement| Added support for `pd.NA` in :class:`impute.SimpleImputer`.
:pr:`21114` by :user:`Ying Xiong <yxiong>`.
- |Enhancement| Adds :term:`get_feature_names_out` to
:class:`impute.SimpleImputer`, :class:`impute.KNNImputer`,
:class:`impute.IterativeImputer`, and :class:`impute.MissingIndicator`.
:pr:`21078` by `Thomas Fan`_.
- |API| The `verbose` parameter was deprecated for :class:`impute.SimpleImputer`.
A warning will always be raised upon the removal of empty columns.
:pr:`21448` by :user:`Oleh Kozynets <OlehKSS>` and
:user:`Christian Ritter <chritter>`.
:mod:`sklearn.inspection`
.........................
- |Feature| Add a display to plot the boundary decision of a classifier by
using the method :func:`inspection.DecisionBoundaryDisplay.from_estimator`.
:pr:`16061` by `Thomas Fan`_.
- |Enhancement| In
:meth:`inspection.PartialDependenceDisplay.from_estimator`, allow
`kind` to accept a list of strings to specify which type of
plot to draw for each feature interaction.
:pr:`19438` by :user:`Guillaume Lemaitre <glemaitre>`.
- |Enhancement| :meth:`inspection.PartialDependenceDisplay.from_estimator`,
:meth:`inspection.PartialDependenceDisplay.plot`, and
`inspection.plot_partial_dependence` now support plotting centered
Individual Conditional Expectation (cICE) and centered PDP curves controlled
by setting the parameter `centered`.
:pr:`18310` by :user:`Johannes Elfner <JoElfner>` and
:user:`Guillaume Lemaitre <glemaitre>`.
:mod:`sklearn.isotonic`
.......................
- |Enhancement| Adds :term:`get_feature_names_out` to
:class:`isotonic.IsotonicRegression`.
:pr:`22249` by `Thomas Fan`_.
:mod:`sklearn.kernel_approximation`
...................................
- |Enhancement| Adds :term:`get_feature_names_out` to
:class:`kernel_approximation.AdditiveChi2Sampler`.
:class:`kernel_approximation.Nystroem`,
:class:`kernel_approximation.PolynomialCountSketch`,
:class:`kernel_approximation.RBFSampler`, and
:class:`kernel_approximation.SkewedChi2Sampler`.
:pr:`22137` and :pr:`22694` by `Thomas Fan`_.
:mod:`sklearn.linear_model`
...........................
- |Feature| :class:`linear_model.ElasticNet`, :class:`linear_model.ElasticNetCV`,
:class:`linear_model.Lasso` and :class:`linear_model.LassoCV` support `sample_weight`
for sparse input `X`.
:pr:`22808` by :user:`Christian Lorentzen <lorentzenchr>`.
- |Feature| :class:`linear_model.Ridge` with `solver="lsqr"` now supports to fit sparse
input with `fit_intercept=True`.
:pr:`22950` by :user:`Christian Lorentzen <lorentzenchr>`.
- |Enhancement| :class:`linear_model.QuantileRegressor` support sparse input
for the highs based solvers.
:pr:`21086` by :user:`Venkatachalam Natchiappan <venkyyuvy>`.
In addition, those solvers now use the CSC matrix right from the
beginning which speeds up fitting.
:pr:`22206` by :user:`Christian Lorentzen <lorentzenchr>`.
- |Enhancement| :class:`linear_model.LogisticRegression` is faster for
``solvers="lbfgs"`` and ``solver="newton-cg"``, for binary and in particular for
multiclass problems thanks to the new private loss function module. In the multiclass
case, the memory consumption has also been reduced for these solvers as the target is
now label encoded (mapped to integers) instead of label binarized (one-hot encoded).
The more classes, the larger the benefit.
:pr:`21808`, :pr:`20567` and :pr:`21814` by
:user:`Christian Lorentzen <lorentzenchr>`.
- |Enhancement| :class:`linear_model.GammaRegressor`,
:class:`linear_model.PoissonRegressor` and :class:`linear_model.TweedieRegressor`
are faster for ``solvers="lbfgs"``.
:pr:`22548`, :pr:`21808` and :pr:`20567` by
:user:`Christian Lorentzen <lorentzenchr>`.
- |Enhancement| Rename parameter `base_estimator` to `estimator` in
:class:`linear_model.RANSACRegressor` to improve readability and consistency.
`base_estimator` is deprecated and will be removed in 1.3.
:pr:`22062` by :user:`Adrian Trujillo <trujillo9616>`.
- |Enhancement| :func:`linear_model.ElasticNet` and
and other linear model classes using coordinate descent show error
messages when non-finite parameter weights are produced. :pr:`22148`
by :user:`Christian Ritter <chritter>` and :user:`Norbert Preining <norbusan>`.
- |Enhancement| :class:`linear_model.ElasticNet` and :class:`linear_model.Lasso`
now raise consistent error messages when passed invalid values for `l1_ratio`,
`alpha`, `max_iter` and `tol`.
:pr:`22240` by :user:`Arturo Amor <ArturoAmorQ>`.
- |Enhancement| :class:`linear_model.BayesianRidge` and
:class:`linear_model.ARDRegression` now preserve float32 dtype. :pr:`9087` by
:user:`Arthur Imbert <Henley13>` and :pr:`22525` by :user:`Meekail Zain <micky774>`.
- |Enhancement| :class:`linear_model.RidgeClassifier` is now supporting
multilabel classification.
:pr:`19689` by :user:`Guillaume Lemaitre <glemaitre>`.
- |Enhancement| :class:`linear_model.RidgeCV` and
:class:`linear_model.RidgeClassifierCV` now raise consistent error message
when passed invalid values for `alphas`.
:pr:`21606` by :user:`Arturo Amor <ArturoAmorQ>`.
- |Enhancement| :class:`linear_model.Ridge` and :class:`linear_model.RidgeClassifier`
now raise consistent error message when passed invalid values for `alpha`,
`max_iter` and `tol`.
:pr:`21341` by :user:`Arturo Amor <ArturoAmorQ>`.
- |Enhancement| :func:`linear_model.orthogonal_mp_gram` preservse dtype for
`numpy.float32`.
:pr:`22002` by :user:`Takeshi Oura <takoika>`.
- |Fix| :class:`linear_model.LassoLarsIC` now correctly computes AIC
and BIC. An error is now raised when `n_features > n_samples` and
when the noise variance is not provided.
:pr:`21481` by :user:`Guillaume Lemaitre <glemaitre>` and
:user:`Andrés Babino <ababino>`.
- |Fix| :class:`linear_model.TheilSenRegressor` now validates input parameter
``max_subpopulation`` in `fit` instead of `__init__`.
:pr:`21767` by :user:`Maren Westermann <marenwestermann>`.
- |Fix| :class:`linear_model.ElasticNetCV` now produces correct
warning when `l1_ratio=0`.
:pr:`21724` by :user:`Yar Khine Phyo <yarkhinephyo>`.
- |Fix| :class:`linear_model.LogisticRegression` and
:class:`linear_model.LogisticRegressionCV` now set the `n_iter_` attribute
with a shape that respects the docstring and that is consistent with the shape
obtained when using the other solvers in the one-vs-rest setting. Previously,
it would record only the maximum of the number of iterations for each binary
sub-problem while now all of them are recorded. :pr:`21998` by
:user:`Olivier Grisel <ogrisel>`.
- |Fix| The property `family` of :class:`linear_model.TweedieRegressor` is not
validated in `__init__` anymore. Instead, this (private) property is deprecated in
:class:`linear_model.GammaRegressor`, :class:`linear_model.PoissonRegressor` and
:class:`linear_model.TweedieRegressor`, and will be removed in 1.3.
:pr:`22548` by :user:`Christian Lorentzen <lorentzenchr>`.
- |Fix| The `coef_` and `intercept_` attributes of
:class:`linear_model.LinearRegression` are now correctly computed in the presence of
sample weights when the input is sparse.
:pr:`22891` by :user:`Jérémie du Boisberranger <jeremiedbb>`.
- |Fix| The `coef_` and `intercept_` attributes of :class:`linear_model.Ridge` with
`solver="sparse_cg"` and `solver="lbfgs"` are now correctly computed in the presence
of sample weights when the input is sparse.
:pr:`22899` by :user:`Jérémie du Boisberranger <jeremiedbb>`.
- |Fix| :class:`linear_model.SGDRegressor` and :class:`linear_model.SGDClassifier` now
computes the validation error correctly when early stopping is enabled.
:pr:`23256` by :user:`Zhehao Liu <MaxwellLZH>`.
- |API| :class:`linear_model.LassoLarsIC` now exposes `noise_variance` as
a parameter in order to provide an estimate of the noise variance.
This is particularly relevant when `n_features > n_samples` and the
estimator of the noise variance cannot be computed.
:pr:`21481` by :user:`Guillaume Lemaitre <glemaitre>`.
:mod:`sklearn.manifold`
.......................
- |Feature| :class:`manifold.Isomap` now supports radius-based
neighbors via the `radius` argument.
:pr:`19794` by :user:`Zhehao Liu <MaxwellLZH>`.
- |Enhancement| :func:`manifold.spectral_embedding` and
:class:`manifold.SpectralEmbedding` supports `np.float32` dtype and will
preserve this dtype.
:pr:`21534` by :user:`Andrew Knyazev <lobpcg>`.
- |Enhancement| Adds :term:`get_feature_names_out` to :class:`manifold.Isomap`
and :class:`manifold.LocallyLinearEmbedding`. :pr:`22254` by `Thomas Fan`_.
- |Enhancement| added `metric_params` to :class:`manifold.TSNE` constructor for
additional parameters of distance metric to use in optimization.
:pr:`21805` by :user:`Jeanne Dionisi <jeannedionisi>` and :pr:`22685` by
:user:`Meekail Zain <micky774>`.
- |Enhancement| :func:`manifold.trustworthiness` raises an error if
`n_neighbours >= n_samples / 2` to ensure a correct support for the function.
:pr:`18832` by :user:`Hong Shao Yang <hongshaoyang>` and :pr:`23033` by
:user:`Meekail Zain <micky774>`.
- |Fix| :func:`manifold.spectral_embedding` now uses Gaussian instead of
the previous uniform on [0, 1] random initial approximations to eigenvectors
in eigen_solvers `lobpcg` and `amg` to improve their numerical stability.
:pr:`21565` by :user:`Andrew Knyazev <lobpcg>`.
:mod:`sklearn.metrics`
......................
- |Feature| :func:`metrics.r2_score` and :func:`metrics.explained_variance_score` have a
new `force_finite` parameter. Setting this parameter to `False` will return the
actual non-finite score in case of perfect predictions or constant `y_true`,
instead of the finite approximation (`1.0` and `0.0` respectively) currently
returned by default. :pr:`17266` by :user:`Sylvain Marié <smarie>`.
- |Feature| :func:`metrics.d2_pinball_score` and :func:`metrics.d2_absolute_error_score`
calculate the :math:`D^2` regression score for the pinball loss and the
absolute error respectively. :func:`metrics.d2_absolute_error_score` is a special case
of :func:`metrics.d2_pinball_score` with a fixed quantile parameter `alpha=0.5`
for ease of use and discovery. The :math:`D^2` scores are generalizations
of the `r2_score` and can be interpreted as the fraction of deviance explained.
:pr:`22118` by :user:`Ohad Michel <ohadmich>`.
- |Enhancement| :func:`metrics.top_k_accuracy_score` raises an improved error
message when `y_true` is binary and `y_score` is 2d. :pr:`22284` by `Thomas Fan`_.
- |Enhancement| :func:`metrics.roc_auc_score` now supports ``average=None``
in the multiclass case when ``multiclass='ovr'`` which will return the score
per class. :pr:`19158` by :user:`Nicki Skafte <SkafteNicki>`.
- |Enhancement| Adds `im_kw` parameter to
:meth:`metrics.ConfusionMatrixDisplay.from_estimator`
:meth:`metrics.ConfusionMatrixDisplay.from_predictions`, and
:meth:`metrics.ConfusionMatrixDisplay.plot`. The `im_kw` parameter is passed
to the `matplotlib.pyplot.imshow` call when plotting the confusion matrix.
:pr:`20753` by `Thomas Fan`_.
- |Fix| :func:`metrics.silhouette_score` now supports integer input for precomputed
distances. :pr:`22108` by `Thomas Fan`_.
- |Fix| Fixed a bug in :func:`metrics.normalized_mutual_info_score` which could return
unbounded values. :pr:`22635` by :user:`Jérémie du Boisberranger <jeremiedbb>`.
- |Fix| Fixes :func:`metrics.precision_recall_curve` and
:func:`metrics.average_precision_score` when true labels are all negative.
:pr:`19085` by :user:`Varun Agrawal <varunagrawal>`.
- |API| `metrics.SCORERS` is now deprecated and will be removed in 1.3. Please
use :func:`metrics.get_scorer_names` to retrieve the names of all available
scorers. :pr:`22866` by `Adrin Jalali`_.
- |API| Parameters ``sample_weight`` and ``multioutput`` of
:func:`metrics.mean_absolute_percentage_error` are now keyword-only, in accordance
with `SLEP009 <https://scikit-learn-enhancement-proposals.readthedocs.io/en/latest/slep009/proposal.html>`_.
A deprecation cycle was introduced.
:pr:`21576` by :user:`Paul-Emile Dugnat <pedugnat>`.
- |API| The `"wminkowski"` metric of :class:`metrics.DistanceMetric` is deprecated
and will be removed in version 1.3. Instead the existing `"minkowski"` metric now takes
in an optional `w` parameter for weights. This deprecation aims at remaining consistent
with SciPy 1.8 convention. :pr:`21873` by :user:`Yar Khine Phyo <yarkhinephyo>`.
- |API| :class:`metrics.DistanceMetric` has been moved from
:mod:`sklearn.neighbors` to :mod:`sklearn.metrics`.
Using `neighbors.DistanceMetric` for imports is still valid for
backward compatibility, but this alias will be removed in 1.3.
:pr:`21177` by :user:`Julien Jerphanion <jjerphan>`.
:mod:`sklearn.mixture`
......................
- |Enhancement| :class:`mixture.GaussianMixture` and
:class:`mixture.BayesianGaussianMixture` can now be initialized using
k-means++ and random data points. :pr:`20408` by
:user:`Gordon Walsh <g-walsh>`, :user:`Alberto Ceballos<alceballosa>`
and :user:`Andres Rios<ariosramirez>`.
- |Fix| Fix a bug that correctly initialize `precisions_cholesky_` in
:class:`mixture.GaussianMixture` when providing `precisions_init` by taking
its square root.
:pr:`22058` by :user:`Guillaume Lemaitre <glemaitre>`.
- |Fix| :class:`mixture.GaussianMixture` now normalizes `weights_` more safely,
preventing rounding errors when calling :meth:`mixture.GaussianMixture.sample` with
`n_components=1`.
:pr:`23034` by :user:`Meekail Zain <micky774>`.
:mod:`sklearn.model_selection`
..............................
- |Enhancement| it is now possible to pass `scoring="matthews_corrcoef"` to all
model selection tools with a `scoring` argument to use the Matthews
correlation coefficient (MCC).
:pr:`22203` by :user:`Olivier Grisel <ogrisel>`.
- |Enhancement| raise an error during cross-validation when the fits for all the
splits failed. Similarly raise an error during grid-search when the fits for
all the models and all the splits failed.
:pr:`21026` by :user:`Loïc Estève <lesteve>`.
- |Fix| :class:`model_selection.GridSearchCV`,
:class:`model_selection.HalvingGridSearchCV`
now validate input parameters in `fit` instead of `__init__`.
:pr:`21880` by :user:`Mrinal Tyagi <MrinalTyagi>`.
- |Fix| :func:`model_selection.learning_curve` now supports `partial_fit`
with regressors. :pr:`22982` by `Thomas Fan`_.
:mod:`sklearn.multiclass`
.........................
- |Enhancement| :class:`multiclass.OneVsRestClassifier` now supports a `verbose`
parameter so progress on fitting can be seen.
:pr:`22508` by :user:`Chris Combs <combscCode>`.
- |Fix| :meth:`multiclass.OneVsOneClassifier.predict` returns correct predictions when
the inner classifier only has a :term:`predict_proba`. :pr:`22604` by `Thomas Fan`_.
:mod:`sklearn.neighbors`
........................
- |Enhancement| Adds :term:`get_feature_names_out` to
:class:`neighbors.RadiusNeighborsTransformer`,
:class:`neighbors.KNeighborsTransformer`
and :class:`neighbors.NeighborhoodComponentsAnalysis`.
:pr:`22212` by :user:`Meekail Zain <micky774>`.
- |Fix| :class:`neighbors.KernelDensity` now validates input parameters in `fit`
instead of `__init__`. :pr:`21430` by :user:`Desislava Vasileva <DessyVV>` and
:user:`Lucy Jimenez <LucyJimenez>`.
- |Fix| :func:`neighbors.KNeighborsRegressor.predict` now works properly when
given an array-like input if `KNeighborsRegressor` is first constructed with a
callable passed to the `weights` parameter. :pr:`22687` by
:user:`Meekail Zain <micky774>`.
:mod:`sklearn.neural_network`
.............................
- |Enhancement| :func:`neural_network.MLPClassifier` and
:func:`neural_network.MLPRegressor` show error
messages when optimizers produce non-finite parameter weights. :pr:`22150`
by :user:`Christian Ritter <chritter>` and :user:`Norbert Preining <norbusan>`.
- |Enhancement| Adds :term:`get_feature_names_out` to
:class:`neural_network.BernoulliRBM`. :pr:`22248` by `Thomas Fan`_.
:mod:`sklearn.pipeline`
.......................
- |Enhancement| Added support for "passthrough" in :class:`pipeline.FeatureUnion`.
Setting a transformer to "passthrough" will pass the features unchanged.
:pr:`20860` by :user:`Shubhraneel Pal <shubhraneel>`.
- |Fix| :class:`pipeline.Pipeline` now does not validate hyper-parameters in
`__init__` but in `.fit()`.
:pr:`21888` by :user:`iofall <iofall>` and :user:`Arisa Y. <arisayosh>`.
- |Fix| :class:`pipeline.FeatureUnion` does not validate hyper-parameters in
`__init__`. Validation is now handled in `.fit()` and `.fit_transform()`.
:pr:`21954` by :user:`iofall <iofall>` and :user:`Arisa Y. <arisayosh>`.
- |Fix| Defines `__sklearn_is_fitted__` in :class:`pipeline.FeatureUnion` to
return correct result with :func:`utils.validation.check_is_fitted`.
:pr:`22953` by :user:`randomgeek78 <randomgeek78>`.
:mod:`sklearn.preprocessing`
............................
- |Feature| :class:`preprocessing.OneHotEncoder` now supports grouping
infrequent categories into a single feature. Grouping infrequent categories
is enabled by specifying how to select infrequent categories with
`min_frequency` or `max_categories`. :pr:`16018` by `Thomas Fan`_.
- |Enhancement| Adds a `subsample` parameter to :class:`preprocessing.KBinsDiscretizer`.
This allows specifying a maximum number of samples to be used while fitting
the model. The option is only available when `strategy` is set to `quantile`.
:pr:`21445` by :user:`Felipe Bidu <fbidu>` and :user:`Amanda Dsouza <amy12xx>`.
- |Enhancement| Adds `encoded_missing_value` to :class:`preprocessing.OrdinalEncoder`
to configure the encoded value for missing data. :pr:`21988` by `Thomas Fan`_.
- |Enhancement| Added the `get_feature_names_out` method and a new parameter
`feature_names_out` to :class:`preprocessing.FunctionTransformer`. You can set
`feature_names_out` to 'one-to-one' to use the input features names as the
output feature names, or you can set it to a callable that returns the output
feature names. This is especially useful when the transformer changes the
number of features. If `feature_names_out` is None (which is the default),
then `get_output_feature_names` is not defined.
:pr:`21569` by :user:`Aurélien Geron <ageron>`.
- |Enhancement| Adds :term:`get_feature_names_out` to
:class:`preprocessing.Normalizer`,
:class:`preprocessing.KernelCenterer`,
:class:`preprocessing.OrdinalEncoder`, and
:class:`preprocessing.Binarizer`. :pr:`21079` by `Thomas Fan`_.
- |Fix| :class:`preprocessing.PowerTransformer` with `method='yeo-johnson'`
better supports significantly non-Gaussian data when searching for an optimal
lambda. :pr:`20653` by `Thomas Fan`_.
- |Fix| :class:`preprocessing.LabelBinarizer` now validates input parameters in
`fit` instead of `__init__`.
:pr:`21434` by :user:`Krum Arnaudov <krumeto>`.
- |Fix| :class:`preprocessing.FunctionTransformer` with `check_inverse=True`
now provides informative error message when input has mixed dtypes. :pr:`19916` by
:user:`Zhehao Liu <MaxwellLZH>`.
- |Fix| :class:`preprocessing.KBinsDiscretizer` handles bin edges more consistently now.
:pr:`14975` by `Andreas Müller`_ and :pr:`22526` by :user:`Meekail Zain <micky774>`.
- |Fix| Adds :meth:`preprocessing.KBinsDiscretizer.get_feature_names_out` support when
`encode="ordinal"`. :pr:`22735` by `Thomas Fan`_.
:mod:`sklearn.random_projection`
................................
- |Enhancement| Adds an `inverse_transform` method and a `compute_inverse_transform`
parameter to :class:`random_projection.GaussianRandomProjection` and
:class:`random_projection.SparseRandomProjection`. When the parameter is set
to True, the pseudo-inverse of the components is computed during `fit` and stored as
`inverse_components_`. :pr:`21701` by :user:`Aurélien Geron <ageron>`.
- |Enhancement| :class:`random_projection.SparseRandomProjection` and
:class:`random_projection.GaussianRandomProjection` preserves dtype for
`numpy.float32`. :pr:`22114` by :user:`Takeshi Oura <takoika>`.
- |Enhancement| Adds :term:`get_feature_names_out` to all transformers in the
:mod:`sklearn.random_projection` module:
:class:`random_projection.GaussianRandomProjection` and
:class:`random_projection.SparseRandomProjection`. :pr:`21330` by
:user:`Loïc Estève <lesteve>`.
:mod:`sklearn.svm`
..................
- |Enhancement| :class:`svm.OneClassSVM`, :class:`svm.NuSVC`,
:class:`svm.NuSVR`, :class:`svm.SVC` and :class:`svm.SVR` now expose
`n_iter_`, the number of iterations of the libsvm optimization routine.
:pr:`21408` by :user:`Juan Martín Loyola <jmloyola>`.
- |Enhancement| :func:`svm.SVR`, :func:`svm.SVC`, :func:`svm.NuSVR`,
:func:`svm.OneClassSVM`, :func:`svm.NuSVC` now raise an error
when the dual-gap estimation produce non-finite parameter weights.
:pr:`22149` by :user:`Christian Ritter <chritter>` and
:user:`Norbert Preining <norbusan>`.
- |Fix| :class:`svm.NuSVC`, :class:`svm.NuSVR`, :class:`svm.SVC`,
:class:`svm.SVR`, :class:`svm.OneClassSVM` now validate input
parameters in `fit` instead of `__init__`.
:pr:`21436` by :user:`Haidar Almubarak <Haidar13 >`.
:mod:`sklearn.tree`
...................
- |Enhancement| :class:`tree.DecisionTreeClassifier` and
:class:`tree.ExtraTreeClassifier` have the new `criterion="log_loss"`, which is
equivalent to `criterion="entropy"`.
:pr:`23047` by :user:`Christian Lorentzen <lorentzenchr>`.
- |Fix| Fix a bug in the Poisson splitting criterion for
:class:`tree.DecisionTreeRegressor`.
:pr:`22191` by :user:`Christian Lorentzen <lorentzenchr>`.
- |API| Changed the default value of `max_features` to 1.0 for
:class:`tree.ExtraTreeRegressor` and to `"sqrt"` for
:class:`tree.ExtraTreeClassifier`, which will not change the fit result. The original
default value `"auto"` has been deprecated and will be removed in version 1.3.
Setting `max_features` to `"auto"` is also deprecated
for :class:`tree.DecisionTreeClassifier` and :class:`tree.DecisionTreeRegressor`.
:pr:`22476` by :user:`Zhehao Liu <MaxwellLZH>`.
:mod:`sklearn.utils`
....................
- |Enhancement| :func:`utils.check_array` and
:func:`utils.multiclass.type_of_target` now accept an `input_name` parameter to make
the error message more informative when passed invalid input data (e.g. with NaN or
infinite values).
:pr:`21219` by :user:`Olivier Grisel <ogrisel>`.
- |Enhancement| :func:`utils.check_array` returns a float
ndarray with `np.nan` when passed a `Float32` or `Float64` pandas extension
array with `pd.NA`. :pr:`21278` by `Thomas Fan`_.
- |Enhancement| :func:`utils.estimator_html_repr` shows a more helpful error
message when running in a jupyter notebook that is not trusted. :pr:`21316`
by `Thomas Fan`_.
- |Enhancement| :func:`utils.estimator_html_repr` displays an arrow on the top
left corner of the HTML representation to show how the elements are
clickable. :pr:`21298` by `Thomas Fan`_.
- |Enhancement| :func:`utils.check_array` with `dtype=None` returns numeric
arrays when passed in a pandas DataFrame with mixed dtypes. `dtype="numeric"`
will also make better infer the dtype when the DataFrame has mixed dtypes.
:pr:`22237` by `Thomas Fan`_.
- |Enhancement| :func:`utils.check_scalar` now has better messages
when displaying the type. :pr:`22218` by `Thomas Fan`_.
- |Fix| Changes the error message of the `ValidationError` raised by
:func:`utils.check_X_y` when y is None so that it is compatible
with the `check_requires_y_none` estimator check. :pr:`22578` by
:user:`Claudio Salvatore Arcidiacono <ClaudioSalvatoreArcidiacono>`.
- |Fix| :func:`utils.class_weight.compute_class_weight` now only requires that
all classes in `y` have a weight in `class_weight`. An error is still raised
when a class is present in `y` but not in `class_weight`. :pr:`22595` by
`Thomas Fan`_.
- |Fix| :func:`utils.estimator_html_repr` has an improved visualization for nested
meta-estimators. :pr:`21310` by `Thomas Fan`_.
- |Fix| :func:`utils.check_scalar` raises an error when
`include_boundaries={"left", "right"}` and the boundaries are not set.
:pr:`22027` by :user:`Marie Lanternier <mlant>`.
- |Fix| :func:`utils.metaestimators.available_if` correctly returns a bounded
method that can be pickled. :pr:`23077` by `Thomas Fan`_.
- |API| :func:`utils.estimator_checks.check_estimator`'s argument is now called
`estimator` (previous name was `Estimator`). :pr:`22188` by
:user:`Mathurin Massias <mathurinm>`.
- |API| ``utils.metaestimators.if_delegate_has_method`` is deprecated and will be
removed in version 1.3. Use :func:`utils.metaestimators.available_if` instead.
:pr:`22830` by :user:`Jérémie du Boisberranger <jeremiedbb>`.
.. rubric:: Code and documentation contributors
Thanks to everyone who has contributed to the maintenance and improvement of
the project since version 1.0, including:
2357juan, Abhishek Gupta, adamgonzo, Adam Li, adijohar, Aditya Kumawat, Aditya
Raghuwanshi, Aditya Singh, Adrian Trujillo Duron, Adrin Jalali, ahmadjubair33,
AJ Druck, aj-white, Alan Peixinho, Alberto Mario Ceballos-Arroyo, Alek
Lefebvre, Alex, Alexandr, Alexandre Gramfort, alexanmv, almeidayoel, Amanda
Dsouza, Aman Sharma, Amar pratap singh, Amit, amrcode, András Simon, Andreas
Grivas, Andreas Mueller, Andrew Knyazev, Andriy, Angus L'Herrou, Ankit Sharma,
Anne Ducout, Arisa, Arth, arthurmello, Arturo Amor, ArturoAmor, Atharva Patil,
aufarkari, Aurélien Geron, avm19, Ayan Bag, baam, Bardiya Ak, Behrouz B,
Ben3940, Benjamin Bossan, Bharat Raghunathan, Bijil Subhash, bmreiniger,
Brandon Truth, Brenden Kadota, Brian Sun, cdrig, Chalmer Lowe, Chiara Marmo,
Chitteti Srinath Reddy, Chloe-Agathe Azencott, Christian Lorentzen, Christian
Ritter, christopherlim98, Christoph T. Weidemann, Christos Aridas, Claudio
Salvatore Arcidiacono, combscCode, Daniela Fernandes, darioka, Darren Nguyen,
Dave Eargle, David Gilbertson, David Poznik, Dea María Léon, Dennis Osei,
DessyVV, Dev514, Dimitri Papadopoulos Orfanos, Diwakar Gupta, Dr. Felix M.
Riese, drskd, Emiko Sano, Emmanouil Gionanidis, EricEllwanger, Erich Schubert,
Eric Larson, Eric Ndirangu, ErmolaevPA, Estefania Barreto-Ojeda, eyast, Fatima
GASMI, Federico Luna, Felix Glushchenkov, fkaren27, Fortune Uwha, FPGAwesome,
francoisgoupil, Frans Larsson, ftorres16, Gabor Berei, Gabor Kertesz, Gabriel
Stefanini Vicente, Gabriel S Vicente, Gael Varoquaux, GAURAV CHOUDHARY,
Gauthier I, genvalen, Geoffrey-Paris, Giancarlo Pablo, glennfrutiz, gpapadok,
Guillaume Lemaitre, Guillermo Tomás Fernández Martín, Gustavo Oliveira, Haidar
Almubarak, Hannah Bohle, Hansin Ahuja, Haoyin Xu, Haya, Helder Geovane Gomes de
Lima, henrymooresc, Hideaki Imamura, Himanshu Kumar, Hind-M, hmasdev, hvassard,
i-aki-y, iasoon, Inclusive Coding Bot, Ingela, iofall, Ishan Kumar, Jack Liu,
Jake Cowton, jalexand3r, J Alexander, Jauhar, Jaya Surya Kommireddy, Jay
Stanley, Jeff Hale, je-kr, JElfner, Jenny Vo, Jérémie du Boisberranger, Jihane,
Jirka Borovec, Joel Nothman, Jon Haitz Legarreta Gorroño, Jordan Silke, Jorge
Ciprián, Jorge Loayza, Joseph Chazalon, Joseph Schwartz-Messing, Jovan
Stojanovic, JSchuerz, Juan Carlos Alfaro Jiménez, Juan Martin Loyola, Julien
Jerphanion, katotten, Kaushik Roy Chowdhury, Ken4git, Kenneth Prabakaran,
kernc, Kevin Doucet, KimAYoung, Koushik Joshi, Kranthi Sedamaki, krishna kumar,
krumetoft, lesnee, Lisa Casino, Logan Thomas, Loic Esteve, Louis Wagner,
LucieClair, Lucy Liu, Luiz Eduardo Amaral, Magali, MaggieChege, Mai,
mandjevant, Mandy Gu, Manimaran, MarcoM, Marco Wurps, Maren Westermann, Maria
Boerner, MarieS-WiMLDS, Martel Corentin, martin-kokos, mathurinm, Matías,
matjansen, Matteo Francia, Maxwell, Meekail Zain, Megabyte, Mehrdad
Moradizadeh, melemo2, Michael I Chen, michalkrawczyk, Micky774, milana2,
millawell, Ming-Yang Ho, Mitzi, miwojc, Mizuki, mlant, Mohamed Haseeb, Mohit
Sharma, Moonkyung94, mpoemsl, MrinalTyagi, Mr. Leu, msabatier, murata-yu, N,
Nadirhan Şahin, Naipawat Poolsawat, NartayXD, nastegiano, nathansquan,
nat-salt, Nicki Skafte Detlefsen, Nicolas Hug, Niket Jain, Nikhil Suresh,
Nikita Titov, Nikolay Kondratyev, Ohad Michel, Oleksandr Husak, Olivier Grisel,
partev, Patrick Ferreira, Paul, pelennor, PierreAttard, Piet Brömmel, Pieter
Gijsbers, Pinky, poloso, Pramod Anantharam, puhuk, Purna Chandra Mansingh,
QuadV, Rahil Parikh, Randall Boyes, randomgeek78, Raz Hoshia, Reshama Shaikh,
Ricardo Ferreira, Richard Taylor, Rileran, Rishabh, Robin Thibaut, Rocco Meli,
Roman Feldbauer, Roman Yurchak, Ross Barnowski, rsnegrin, Sachin Yadav,
sakinaOuisrani, Sam Adam Day, Sanjay Marreddi, Sebastian Pujalte, SEELE, SELEE,
Seyedsaman (Sam) Emami, ShanDeng123, Shao Yang Hong, sharmadharmpal,
shaymerNaturalint, Shuangchi He, Shubhraneel Pal, siavrez, slishak, Smile,
spikebh, sply88, Srinath Kailasa, Stéphane Collot, Sultan Orazbayev, Sumit
Saha, Sven Eschlbeck, Sven Stehle, Swapnil Jha, Sylvain Marié, Takeshi Oura,
Tamires Santana, Tenavi, teunpe, Theis Ferré Hjortkjær, Thiruvenkadam, Thomas
J. Fan, t-jakubek, toastedyeast, Tom Dupré la Tour, Tom McTiernan, TONY GEORGE,
Tyler Martin, Tyler Reddy, Udit Gupta, Ugo Marchand, Varun Agrawal,
Venkatachalam N, Vera Komeyer, victoirelouis, Vikas Vishwakarma, Vikrant
khedkar, Vladimir Chernyy, Vladimir Kim, WeijiaDu, Xiao Yuan, Yar Khine Phyo,
Ying Xiong, yiyangq, Yosshi999, Yuki Koyama, Zach Deane-Mayer, Zeel B Patel,
zempleni, zhenfisher, 赵丰 (Zhao Feng)