8000 FEA metadata routing for `StackingClassifier` and `StackingRegressor`… · scikit-learn/scikit-learn@61281cf · GitHub
[go: up one dir, main page]
Skip to content

Commit 61281cf

Browse files
StefanieSengeradrinjalaliOmarManzoor
authored
FEA metadata routing for StackingClassifier and StackingRegressor (#28701)
Co-authored-by: Adrin Jalali <adrin.jalali@gmail.com> Co-authored-by: Omar Salman <omar.salman@arbisoft.com>
1 parent 2d6e1be commit 61281cf

File tree

8 files changed

+335
-56
lines changed

8 files changed

+335
-56
lines changed

‎doc/metadata_routing.rst

Lines changed: 2 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -277,6 +277,8 @@ Meta-estimators and functions supporting metadata routing:
277277
- :class:`sklearn.calibration.CalibratedClassifierCV`
278278
- :class:`sklearn.compose.ColumnTransformer`
279279
- :class:`sklearn.covariance.GraphicalLassoCV`
280+
- :class:`sklearn.ensemble.StackingClassifier`
281+
- :class:`sklearn.ensemble.StackingRegressor`
280282
- :class:`sklearn.ensemble.VotingClassifier`
281283
- :class:`sklearn.ensemble.VotingRegressor`
282284
- :class:`sklearn.ensemble.BaggingClassifier`
@@ -316,13 +318,9 @@ Meta-estimators and tools not supporting metadata routing yet:
316318
- :class:`sklearn.compose.TransformedTargetRegressor`
317319
- :class:`sklearn.ensemble.AdaBoostClassifier`
318320
- :class:`sklearn.ensemble.AdaBoostRegressor`
319-
- :class:`sklearn.ensemble.StackingClassifier`
320-
- :class:`sklearn.ensemble.StackingRegressor`
321321
- :class:`sklearn.feature_selection.RFE`
322322
- :class:`sklearn.feature_selection.RFECV`
323323
- :class:`sklearn.feature_selection.SequentialFeatureSelector`
324-
- :class:`sklearn.impute.IterativeImputer`
325-
- :class:`sklearn.linear_model.RANSACRegressor`
326324
- :class:`sklearn.model_selection.learning_curve`
327325
- :class:`sklearn.model_selection.permutation_test_score`
328326
- :class:`sklearn.model_selection.validation_curve`

‎doc/modules/ensemble.rst

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1581,8 +1581,8 @@ availability, tested in the order of preference: `predict_proba`,
15811581
`decision_function` and `predict`.
15821582

15831583
A :class:`StackingRegressor` and :class:`StackingClassifier` can be used as
1584-
any other regressor or classifier, exposing a `predict`, `predict_proba`, and
1585-
`decision_function` methods, e.g.::
1584+
any other regressor or classifier, exposing a `predict`, `predict_proba`, or
1585+
`decision_function` method, e.g.::
15861586

15871587
>>> y_pred = reg.predict(X_test)
15881588
>>> from sklearn.metrics import r2_score

‎doc/whats_new/v1.6.rst

Lines changed: 13 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -38,7 +38,19 @@ See :ref:`array_api` for more details.
3838

3939
**Classes:**
4040

41-
-
41+
-
42+
43+
Metadata Routing
44+
----------------
45+
46+
The following models now support metadata routing in one or more of their
47+
methods. Refer to the :ref:`Metadata Routing User Guide <metadata_routing>` for
48+
more details.
49+
50+
- |Feature| :class:`ensemble.StackingClassifier` and
51+
:class:`ensemble.StackingRegressor` now support metadata routing and pass
52+
``**fit_params`` to the underlying estimators via their `fit` methods.
53+
:pr:`28701` by :user:`Stefanie Senger <StefanieSenger>`.
4254

4355
Changelog
4456
---------

‎sklearn/ensemble/_base.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -21,7 +21,7 @@ def _fit_single_estimator(
2121
estimator, X, y, fit_params, message_clsname=None, message=None
2222
):
2323
"""Private function used to fit an estimator within a job."""
24-
# TODO(SLEP6): remove if condition for unrouted sample_weight when metadata
24+
# TODO(SLEP6): remove if-condition for unrouted sample_weight when metadata
2525
# routing can't be disabled.
2626
if not _routing_enabled() and "sample_weight" in fit_params:
2727
try:

0 commit comments

Comments
 (0)
0