scikit-learn · adrinjalali · Sep 14, 2021 · Sep 7, 2021 · Sep 7, 2021 · Sep 7, 2021
diff --git a/.circleci/config.yml b/.circleci/config.yml
@@ -187,11 +187,6 @@ workflows:
       - doc-min-dependencies:
           requires:
             - lint
-      - pypy3:
-          filters:
-            branches:
-              only:
-                - 0.20.X
       - deploy:
           requires:
             - doc

diff --git a/.travis.yml b/.travis.yml
@@ -18,15 +18,16 @@ env:
     - OPENBLAS_NUM_THREADS=2
     - SKLEARN_BUILD_PARALLEL=3
     - SKLEARN_SKIP_NETWORK_TESTS=1
+    - PYTHONUNBUFFERED=1
     # Custom environment variables for the ARM wheel builder
     - CIBW_BUILD_VERBOSITY=1
-    - CIBW_TEST_REQUIRES="pytest pytest-xdist threadpoolctl"
     - CIBW_TEST_COMMAND="bash {project}/build_tools/travis/test_wheels.sh"
-    - CIBW_ENVIRONMENT="CPU_COUNT=2
+    - CIBW_ENVIRONMENT="CPU_COUNT=6
                         OMP_NUM_THREADS=2
                         OPENBLAS_NUM_THREADS=2
-                        SKLEARN_BUILD_PARALLEL=3
-                        SKLEARN_SKIP_NETWORK_TESTS=1"
+                        SKLEARN_BUILD_PARALLEL=10
+                        SKLEARN_SKIP_NETWORK_TESTS=1
+                        PYTHONUNBUFFERED=1"
 
 jobs:
   include:

diff --git a/build_tools/travis/install.sh b/build_tools/travis/install.sh
@@ -4,10 +4,8 @@
 # defined in the ".travis.yml" file. In particular, it is
 # important that we call to the right installation script.
 
-set -e
-
 if [[ $BUILD_WHEEL == true ]]; then
-    source build_tools/travis/install_wheels.sh
+    source build_tools/travis/install_wheels.sh || travis_terminate 1
 else
-    source build_tools/travis/install_main.sh
+    source build_tools/travis/install_main.sh || travis_terminate 1
 fi
diff --git a/build_tools/travis/install_wheels.sh b/build_tools/travis/install_wheels.sh
@@ -1,6 +1,4 @@
 #!/bin/bash
 
-set -e
-
-python -m pip install cibuildwheel
-python -m cibuildwheel --output-dir wheelhouse
+python -m pip install cibuildwheel || travis_terminate $?
+python -m cibuildwheel --output-dir wheelhouse || travis_terminate $?
diff --git a/build_tools/travis/script.sh b/build_tools/travis/script.sh
@@ -5,8 +5,6 @@
 # continuous deployment jobs, we have to execute the scripts for
 # testing the continuous integration jobs.
 
-set -e
-
 if [[ $BUILD_WHEEL != true ]]; then
     # This trick will make Travis terminate the continuation of the pipeline
     bash build_tools/travis/test_script.sh || travis_terminate 1

diff --git a/build_tools/travis/test_wheels.sh b/build_tools/travis/test_wheels.sh
@@ -1,9 +1,9 @@
 #!/bin/bash
 
-set -e
+pip install --upgrade pip || travis_terminate $?
+pip install pytest pytest-xdist || travis_terminate $?
 
-# Faster run of the source code tests
-pytest -n $CPU_COUNT --pyargs sklearn
+python -m pytest -n $CPU_COUNT --pyargs sklearn || travis_terminate $?
 
 # Test that there are no links to system libraries
-python -m threadpoolctl -i sklearn
+python -m threadpoolctl -i sklearn || travis_terminate $?
diff --git a/doc/developers/plotting.rst b/doc/developers/plotting.rst
@@ -64,8 +64,8 @@ Plotting with Multiple Axes
 ---------------------------
 
 Some of the plotting tools like
-:func:`~sklearn.inspection.plot_partial_dependence` and
-:class:`~sklearn.inspection.PartialDependenceDisplay` support plottong on
+:func:`~sklearn.inspection.PartialDependenceDisplay.from_estimator` and
+:class:`~sklearn.inspection.PartialDependenceDisplay` support plotting on
 multiple axes. Two different scenarios are supported:
 
 1. If a list of axes is passed in, `plot` will check if the number of axes is
@@ -87,8 +87,8 @@ be placed. In this case, we suggest using matplotlib's
 By default, the `ax` keyword in `plot` is `None`. In this case, the single
 axes is created and the gridspec api is used to create the regions to plot in.
 
-See for example, :func:`~sklearn.inspection.plot_partial_dependence` which
-plots multiple lines and contours using this API. The axes defining the
+See for example, :func:`~sklearn.inspection.PartialDependenceDisplay.from_estimator
+which plots multiple lines and contours using this API. The axes defining the
 bounding box is saved in a `bounding_ax_` attribute. The individual axes
 created are stored in an `axes_` ndarray, corresponding to the axes position on
 the grid. Positions that are not used are set to `None`. Furthermore, the

diff --git a/doc/modules/partial_dependence.rst b/doc/modules/partial_dependence.rst
@@ -55,20 +55,20 @@ independent of the house age, whereas for values less than 2 there is a strong
 dependence on age.
 
 The :mod:`sklearn.inspection` module provides a convenience function
-:func:`plot_partial_dependence` to create one-way and two-way partial
+:func:`~PartialDependenceDisplay.from_estimator` to create one-way and two-way partial
 dependence plots. In the below example we show how to create a grid of
 partial dependence plots: two one-way PDPs for the features ``0`` and ``1``
 and a two-way PDP between the two features::
 
     >>> from sklearn.datasets import make_hastie_10_2
     >>> from sklearn.ensemble import GradientBoostingClassifier
-    >>> from sklearn.inspection import plot_partial_dependence
+    >>> from sklearn.inspection import PartialDependenceDisplay
 
     >>> X, y = make_hastie_10_2(random_state=0)
     >>> clf = GradientBoostingClassifier(n_estimators=100, learning_rate=1.0,
     ...     max_depth=1, random_state=0).fit(X, y)
     >>> features = [0, 1, (0, 1)]
-    >>> plot_partial_dependence(clf, X, features)
+    >>> PartialDependenceDisplay.from_estimator(clf, X, features)
     <...>
 
 You can access the newly created figure and Axes objects using ``plt.gcf()``
@@ -82,7 +82,7 @@ the PDPs should be created via the ``target`` argument::
     >>> mc_clf = GradientBoostingClassifier(n_estimators=10,
     ...     max_depth=1).fit(iris.data, iris.target)
     >>> features = [3, 2, (3, 2)]
-    >>> plot_partial_dependence(mc_clf, X, features, target=0)
+    >>> PartialDependenceDisplay.from_estimator(mc_clf, X, features, target=0)
     <...>
 
 The same parameter ``target`` is used to specify the target in multi-output
@@ -138,20 +138,20 @@ and the house price in the PD line. However, the ICE lines show that there
 are some exceptions, where the house price remains constant in some ranges of
 the median income.
 
-The :mod:`sklearn.inspection` module's :func:`plot_partial_dependence`
+The :mod:`sklearn.inspection` module's :meth:`PartialDependenceDisplay.from_estimator`
 convenience function can be used to create ICE plots by setting
 ``kind='individual'``. In the example below, we show how to create a grid of
 ICE plots:
 
     >>> from sklearn.datasets import make_hastie_10_2
     >>> from sklearn.ensemble import GradientBoostingClassifier
-    >>> from sklearn.inspection import plot_partial_dependence
+    >>> from sklearn.inspection import PartialDependenceDisplay
 
     >>> X, y = make_hastie_10_2(random_state=0)
     >>> clf = GradientBoostingClassifier(n_estimators=100, learning_rate=1.0,
     ...     max_depth=1, random_state=0).fit(X, y)
     >>> features = [0, 1]
-    >>> plot_partial_dependence(clf, X, features,
+    >>> PartialDependenceDisplay.from_estimator(clf, X, features,
     ...     kind='individual')
     <...>
 
@@ -160,7 +160,7 @@ feature of interest. Hence, it is recommended to use ICE plots alongside
 PDPs. They can be plotted together with
 ``kind='both'``.
 
-    >>> plot_partial_dependence(clf, X, features,
+    >>> PartialDependenceDisplay.from_estimator(clf, X, features,
     ...     kind='both')
     <...>
 

diff --git a/doc/whats_new/v1.0.rst b/doc/whats_new/v1.0.rst
@@ -461,6 +461,11 @@ Changelog
   :func:`~sklearn.inspection.permutation_importance`.
   :pr:`19411` by :user:`Simona Maggio <simonamaggio>`.
 
+- |API| :class:`inspection.PartialDependenceDisplay` exposes a class method:
+  :func:`~inspection.PartialDependenceDisplay.from_estimator`.
+  :func:`inspection.plot_partial_dependence` is deprecated in favor of the
+  class method and will be removed in 1.2. :pr:`20959` by `Thomas Fan`_.
+
 :mod:`sklearn.kernel_approximation`
 ...................................
 
@@ -703,6 +708,9 @@ Changelog
 - |Enhancement| warn only once in the main process for per-split fit failures
   in cross-validation. :pr:`20619` by :user:`Loïc Estève <lesteve>`
 
+- |Enhancement| The :class:`model_selection.BaseShuffleSplit` base class is
+  now public. :pr:`20056` by :user:`pabloduque0`.
+
 - |Fix| Avoid premature overflow in :func:`model_selection.train_test_split`.
   :pr:`20904` by :user:`Tomasz Jakubek <t-jakubek>`.
 
@@ -821,6 +829,9 @@ Changelog
   `n_features_in_` and will be removed in 1.2. :pr:`20240` by
   :user:`Jérémie du Boisberranger <jeremiedbb>`.
 
+- |Fix| :class:`preprocessing.FunctionTransformer` does not set `n_features_in_`
+  based on the input to `inverse_transform`. :pr:`20961` by `Thomas Fan`_.
+
 :mod:`sklearn.svm`
 ...................
 
@@ -895,7 +906,7 @@ Changelog
   warning was previously raised in resampling utilities and functions using
   those utilities (e.g. :func:`model_selection.train_test_split`,
   :func:`model_selection.cross_validate`,
-  :func:`model_seleection.cross_val_score`,
+  :func:`model_selection.cross_val_score`,
   :func:`model_selection.cross_val_predict`).
   :pr:`20673` by :user:`Joris Van den Bossche  <jorisvandenbossche>`.
 

diff --git a/examples/ensemble/plot_monotonic_constraints.py b/examples/ensemble/plot_monotonic_constraints.py
@@ -19,7 +19,7 @@
 <https://xgboost.readthedocs.io/en/latest/tutorials/monotonic.html>`_.
 """
 from sklearn.ensemble import HistGradientBoostingRegressor
-from sklearn.inspection import plot_partial_dependence
+from sklearn.inspection import PartialDependenceDisplay
 import numpy as np
 import matplotlib.pyplot as plt
 
@@ -43,7 +43,7 @@
 # Without any constraint
 gbdt = HistGradientBoostingRegressor()
 gbdt.fit(X, y)
-disp = plot_partial_dependence(
+disp = PartialDependenceDisplay.from_estimator(
     gbdt,
     X,
     features=[0, 1],
@@ -55,7 +55,7 @@
 gbdt = HistGradientBoostingRegressor(monotonic_cst=[1, -1])
 gbdt.fit(X, y)
 
-plot_partial_dependence(
+PartialDependenceDisplay.from_estimator(
     gbdt,
     X,
     features=[0, 1],

diff --git a/examples/inspection/plot_partial_dependence.py b/examples/inspection/plot_partial_dependence.py
@@ -111,12 +111,12 @@
 
 import matplotlib.pyplot as plt
 from sklearn.inspection import partial_dependence
-from sklearn.inspection import plot_partial_dependence
+from sklearn.inspection import PartialDependenceDisplay
 
 print("Computing partial dependence plots...")
 tic = time()
 features = ["MedInc", "AveOccup", "HouseAge", "AveRooms"]
-display = plot_partial_dependence(
+display = PartialDependenceDisplay.from_estimator(
     est,
     X_train,
     features,
@@ -166,7 +166,7 @@
 
 print("Computing partial dependence plots...")
 tic = time()
-display = plot_partial_dependence(
+display = PartialDependenceDisplay.from_estimator(
     est,
     X_train,
     features,
@@ -229,7 +229,7 @@
 print("Computing partial dependence plots...")
 tic = time()
 _, ax = plt.subplots(ncols=3, figsize=(9, 4))
-display = plot_partial_dependence(
+display = PartialDependenceDisplay.from_estimator(
     est,
     X_train,
     features,

diff --git a/examples/miscellaneous/plot_partial_dependence_visualization_api.py b/examples/miscellaneous/plot_partial_dependence_visualization_api.py
@@ -22,7 +22,7 @@
 from sklearn.preprocessing import StandardScaler
 from sklearn.pipeline import make_pipeline
 from sklearn.tree import DecisionTreeRegressor
-from sklearn.inspection import plot_partial_dependence
+from sklearn.inspection import PartialDependenceDisplay
 
 
 # %%
@@ -49,22 +49,22 @@
 #
 # We plot partial dependence curves for features "age" and "bmi" (body mass
 # index) for the decision tree. With two features,
-# :func:`~sklearn.inspection.plot_partial_dependence` expects to plot two
-# curves. Here the plot function place a grid of two plots using the space
+# :func:`~sklearn.inspection.PartialDependenceDisplay.from_estimator` expects to plot
+# two curves. Here the plot function place a grid of two plots using the space
 # defined by `ax` .
 fig, ax = plt.subplots(figsize=(12, 6))
 ax.set_title("Decision Tree")
-tree_disp = plot_partial_dependence(tree, X, ["age", "bmi"], ax=ax)
+tree_disp = PartialDependenceDisplay.from_estimator(tree, X, ["age", "bmi"], ax=ax)
 
 # %%
 # The partial depdendence curves can be plotted for the multi-layer perceptron.
 # In this case, `line_kw` is passed to
-# :func:`~sklearn.inspection.plot_partial_dependence` to change the color of
-# the curve.
+# :func:`~sklearn.inspection.PartialDependenceDisplay.from_estimator` to change the
+# color of the curve.
 fig, ax = plt.subplots(figsize=(12, 6))
 ax.set_title("Multi-layer Perceptron")
-mlp_disp = plot_partial_dependence(mlp, X, ["age", "bmi"], ax=ax,
-                                   line_kw={"color": "red"})
+mlp_disp = PartialDependenceDisplay.from_estimator(mlp, X, ["age", "bmi"], ax=ax,
+                                                   line_kw={"color": "red"})
 
 # %%
 # Plotting partial dependence of the two models together
@@ -129,7 +129,6 @@
 # Here, we plot the partial dependence curves for a single feature, "age", on
 # the same axes. In this case, `tree_disp.axes_` is passed into the second
 # plot function.
-tree_disp = plot_partial_dependence(tree, X, ["age"])
-mlp_disp = plot_partial_dependence(mlp, X, ["age"],
-                                   ax=tree_disp.axes_,
-                                   line_kw={"color": "red"})
+tree_disp = PartialDependenceDisplay.from_estimator(tree, X, ["age"])
+mlp_disp = PartialDependenceDisplay.from_estimator(
+    mlp, X, ["age"], ax=tree_disp.axes_, line_kw={"color": "red"})
diff --git a/sklearn/calibration.py b/sklearn/calibration.py
@@ -980,8 +980,8 @@ class CalibrationDisplay:
     y_prob : ndarray of shape (n_samples,)
         Probability estimates for the positive class, for each sample.
 
-    name : str, default=None
-        Name for labeling curve.
+    estimator_name : str, default=None
+        Name of estimator. If None, the estimator name is not shown.
 
     Attributes
     ----------
@@ -1022,11 +1022,11 @@ class CalibrationDisplay:
     <...>
     """
 
-    def __init__(self, prob_true, prob_pred, y_prob, *, name=None):
+    def __init__(self, prob_true, prob_pred, y_prob, *, estimator_name=None):
         self.prob_true = prob_true
         self.prob_pred = prob_pred
         self.y_prob = y_prob
-        self.name = name
+        self.estimator_name = estimator_name
 
     def plot(self, *, ax=None, name=None, ref_line=True, **kwargs):
         """Plot visualization.
@@ -1041,7 +1041,8 @@ def plot(self, *, ax=None, name=None, ref_line=True, **kwargs):
             created.
 
         name : str, default=None
-            Name for labeling curve.
+            Name for labeling curve. If `None`, use `estimator_name` if
+            not `None`, otherwise no labeling is shown.
 
         ref_line : bool, default=True
             If `True`, plots a reference line representing a perfectly
@@ -1061,8 +1062,7 @@ def plot(self, *, ax=None, name=None, ref_line=True, **kwargs):
         if ax is None:
             fig, ax = plt.subplots()
 
-        name = self.name if name is None else name
-        self.name = name
+        name = self.estimator_name if name is None else name
 
         line_kwargs = {}
         if name is not None:
@@ -1298,6 +1298,9 @@ def from_predictions(
         prob_true, prob_pred = calibration_curve(
             y_true, y_prob, n_bins=n_bins, strategy=strategy
         )
+        name = name if name is not None else "Classifier"
 
-        disp = cls(prob_true=prob_true, prob_pred=prob_pred, y_prob=y_prob, name=name)
+        disp = cls(
+            prob_true=prob_true, prob_pred=prob_pred, y_prob=y_prob, estimator_name=name
+        )
         return disp.plot(ax=ax, ref_line=ref_line, **kwargs)
diff --git a/sklearn/cluster/_affinity_propagation.py b/sklearn/cluster/_affinity_propagation.py
@@ -272,7 +272,7 @@ class AffinityPropagation(ClusterMixin, BaseEstimator):
     Parameters
     ----------
     damping : float, default=0.5
-        Damping factor (between 0.5 and 1) is the extent to
+        Damping factor in the range `[0.5, 1.0)` is the extent to
         which the current value is maintained relative to
         incoming values (weighted 1 - damping). This in order
         to avoid numerical oscillations when updating these
@@ -469,7 +469,7 @@ def fit(self, X, y=None):
             target_type=numbers.Real,
             min_val=0.5,
             max_val=1,
-            closed="right",
+            include_boundaries="left",
         )
         check_scalar(self.max_iter, "max_iter", target_type=numbers.Integral, min_val=1)
         check_scalar(

diff --git a/sklearn/inspection/_partial_dependence.py b/sklearn/inspection/_partial_dependence.py
@@ -370,7 +370,7 @@ def partial_dependence(
 
     See Also
     --------
-    plot_partial_dependence : Plot Partial Dependence.
+    PartialDependenceDisplay.from_estimator : Plot Partial Dependence.
     PartialDependenceDisplay : Partial Dependence visualization.
 
     Examples