scikit-learn · amueller · Jul 17, 2018 · Jul 16, 2018 · Jul 16, 2018 · Jul 16, 2018
diff --git a/setup.cfg b/setup.cfg
@@ -8,6 +8,9 @@ addopts =
     --doctest-modules
     --disable-pytest-warnings
     -rs
+filterwarnings =
+    error::DeprecationWarning
+    error::FutureWarning
 
 [wheelhouse_uploader]
 artifact_indexes=

diff --git a/sklearn/cluster/tests/test_hierarchical.py b/sklearn/cluster/tests/test_hierarchical.py
@@ -7,6 +7,7 @@
 # License: BSD 3 clause
 from tempfile import mkdtemp
 import shutil
+import pytest
 from functools import partial
 
 import numpy as np
@@ -142,6 +143,8 @@ def test_agglomerative_clustering_wrong_arg_memory():
     assert_raises(ValueError, clustering.fit, X)
 
 
+@pytest.mark.filterwarnings("ignore:the behavior of nmi will "
+                            "change in version 0.22")
 def test_agglomerative_clustering():
     # Check that we obtain the correct number of clusters with
     # agglomerative clustering.
@@ -250,6 +253,8 @@ def test_ward_agglomeration():
     assert_raises(ValueError, agglo.fit, X[:0])
 
 
+@pytest.mark.filterwarnings("ignore:the behavior of nmi will "
+                            "change in version 0.22")
 def test_single_linkage_clustering():
     # Check that we get the correct result in two emblematic cases
     moons, moon_labels = make_moons(noise=0.05, random_state=42)
@@ -311,6 +316,8 @@ def test_scikit_vs_scipy():
     assert_raises(ValueError, _hc_cut, n_leaves + 1, children, n_leaves)
 
 
+@pytest.mark.filterwarnings("ignore:the behavior of nmi will "
+                            "change in version 0.22")
 def test_identical_points():
     # Ensure identical points are handled correctly when using mst with
     # a sparse connectivity matrix

diff --git a/sklearn/covariance/tests/test_graphical_lasso.py b/sklearn/covariance/tests/test_graphical_lasso.py
@@ -4,6 +4,7 @@
 
 import numpy as np
 from scipy import linalg
+import pytest
 
 from sklearn.utils.testing import assert_array_almost_equal
 from sklearn.utils.testing import assert_array_less
@@ -151,6 +152,5 @@ def test_deprecated_grid_scores(random_state=1):
                     "0.19 and will be removed in 0.21. Use "
                     "``grid_scores_`` instead")
 
-    assert_warns_message(DeprecationWarning, depr_message,
-                         lambda: graphical_lasso.grid_scores)
-    assert_equal(graphical_lasso.grid_scores, graphical_lasso.grid_scores_)
+    with pytest.warns(DeprecationWarning, match=depr_message):
+        assert_equal(graphical_lasso.grid_scores, graphical_lasso.grid_scores_)
diff --git a/sklearn/decomposition/tests/test_dict_learning.py b/sklearn/decomposition/tests/test_dict_learning.py
@@ -58,6 +58,7 @@ def test_dict_learning_overcomplete():
     assert_true(dico.components_.shape == (n_components, n_features))
 
 
+@ignore_warnings(category=DeprecationWarning)  # positive lars deprecated 0.22
 @pytest.mark.parametrize("transform_algorithm", [
     "lasso_lars",
     "lasso_cd",
@@ -170,6 +171,7 @@ def test_dict_learning_online_shapes():
     assert_equal(np.dot(code, dictionary).shape, X.shape)
 
 
+@ignore_warnings(category=DeprecationWarning)  # positive lars deprecated 0.22
 @pytest.mark.parametrize("transform_algorithm", [
     "lasso_lars",
     "lasso_cd",
@@ -306,6 +308,7 @@ def test_sparse_encode_shapes():
         assert_equal(code.shape, (n_samples, n_components))
 
 
+@ignore_warnings(category=DeprecationWarning)  # positive lars deprecated 0.22
 @pytest.mark.parametrize("positive", [
     False,
     True,

diff --git a/sklearn/decomposition/tests/test_kernel_pca.py b/sklearn/decomposition/tests/test_kernel_pca.py
@@ -1,9 +1,10 @@
 import numpy as np
 import scipy.sparse as sp
+import pytest
 
 from sklearn.utils.testing import (assert_array_almost_equal, assert_less,
                                    assert_equal, assert_not_equal,
-                                   assert_raises)
+                                   assert_raises, ignore_warnings)
 
 from sklearn.decomposition import PCA, KernelPCA
 from sklearn.datasets import make_circles
@@ -172,6 +173,7 @@ def test_kernel_pca_invalid_kernel():
     assert_raises(ValueError, kpca.fit, X_fit)
 
 
+@pytest.mark.filterwarnings('ignore: The default of the `iid`')  # 0.22
 def test_gridsearch_pipeline():
     # Test if we can do a grid-search to find parameters to separate
     # circles with a perceptron model.
@@ -186,6 +188,7 @@ def test_gridsearch_pipeline():
     assert_equal(grid_search.best_score_, 1)
 
 
+@pytest.mark.filterwarnings('ignore: The default of the `iid`')  # 0.22
 def test_gridsearch_pipeline_precomputed():
     # Test if we can do a grid-search to find parameters to separate
     # circles with a perceptron model using a precomputed kernel.

diff --git a/sklearn/ensemble/forest.py b/sklearn/ensemble/forest.py
@@ -944,16 +944,17 @@ class labels (multi-output problem).
     >>> X, y = make_classification(n_samples=1000, n_features=4,
     ...                            n_informative=2, n_redundant=0,
     ...                            random_state=0, shuffle=False)
-    >>> clf = RandomForestClassifier(max_depth=2, random_state=0)
+    >>> clf = RandomForestClassifier(n_estimators=100, max_depth=2,
+    ...                              random_state=0)
     >>> clf.fit(X, y)
     RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',
                 max_depth=2, max_features='auto', max_leaf_nodes=None,
                 min_impurity_decrease=0.0, min_impurity_split=None,
                 min_samples_leaf=1, min_samples_split=2,
-                min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,
+                min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=1,
                 oob_score=False, random_state=0, verbose=0, warm_start=False)
     >>> print(clf.feature_importances_)
-    [0.17287856 0.80608704 0.01884792 0.00218648]
+    [0.14205973 0.76664038 0.0282433  0.06305659]
     >>> print(clf.predict([[0, 0, 0, 0]]))
     [1]
 
@@ -1188,18 +1189,19 @@ class RandomForestRegressor(ForestRegressor):
     >>>
     >>> X, y = make_regression(n_features=4, n_informative=2,
     ...                        random_state=0, shuffle=False)
-    >>> regr = RandomForestRegressor(max_depth=2, random_state=0)
+    >>> regr = RandomForestRegressor(max_depth=2, random_state=0,
+    ...                              n_estimators=100)
     >>> regr.fit(X, y)
     RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=2,
                max_features='auto', max_leaf_nodes=None,
                min_impurity_decrease=0.0, min_impurity_split=None,
                min_samples_leaf=1, min_samples_split=2,
-               min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,
+               min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=1,
                oob_score=False, random_state=0, verbose=0, warm_start=False)
     >>> print(regr.feature_importances_)
-    [0.17339552 0.81594114 0.         0.01066333]
+    [0.18146984 0.81473937 0.00145312 0.00233767]
     >>> print(regr.predict([[0, 0, 0, 0]]))
-    [-2.50699856]
+    [-8.32987858]
 
     Notes
     -----

diff --git a/sklearn/ensemble/tests/test_bagging.py b/sklearn/ensemble/tests/test_bagging.py
@@ -5,6 +5,7 @@
 # Author: Gilles Louppe
 # License: BSD 3 clause
 
+import pytest
 import numpy as np
 
 from sklearn.base import BaseEstimator
@@ -33,7 +34,7 @@
 from sklearn.model_selection import train_test_split
 from sklearn.datasets import load_boston, load_iris, make_hastie_10_2
 from sklearn.utils import check_random_state
-from sklearn.preprocessing import Imputer
+from sklearn.preprocessing import FunctionTransformer
 
 from scipy.sparse import csc_matrix, csr_matrix
 
@@ -496,6 +497,7 @@ def test_parallel_regression():
     assert_array_almost_equal(y1, y3)
 
 
+@pytest.mark.filterwarnings('ignore: The default of the `iid`')  # 0.22
 def test_gridsearch():
     # Check that bagging ensembles can be grid-searched.
     # Transform iris into a binary classification task
@@ -755,6 +757,12 @@ def test_set_oob_score_label_encoding():
     assert_equal([x1, x2], [x3, x3])
 
 
+def replace(X):
+    X = X.copy().astype('float')
+    X[~np.isfinite(X)] = 0
+    return X
+
+
 def test_bagging_regressor_with_missing_inputs():
     # Check that BaggingRegressor can accept X with missing/infinite data
     X = np.array([
@@ -777,9 +785,7 @@ def test_bagging_regressor_with_missing_inputs():
     for y in y_values:
         regressor = DecisionTreeRegressor()
         pipeline = make_pipeline(
-            Imputer(),
-            Imputer(missing_values=np.inf),
-            Imputer(missing_values=np.NINF),
+            FunctionTransformer(replace, validate=False),
             regressor
         )
         pipeline.fit(X, y).predict(X)
@@ -807,9 +813,7 @@ def test_bagging_classifier_with_missing_inputs():
     y = np.array([3, 6, 6, 6, 6])
     classifier = DecisionTreeClassifier()
     pipeline = make_pipeline(
-        Imputer(),
-        Imputer(missing_values=np.inf),
-        Imputer(missing_values=np.NINF),
+        FunctionTransformer(replace, validate=False),
         classifier
     )
     pipeline.fit(X, y).predict(X)

diff --git a/sklearn/ensemble/tests/test_forest.py b/sklearn/ensemble/tests/test_forest.py
@@ -439,13 +439,13 @@ def check_oob_score_raise_error(name):
 def test_oob_score_raise_error(name):
     check_oob_score_raise_error(name)
 
-
 def check_gridsearch(name):
     forest = FOREST_CLASSIFIERS[name]()
     clf = GridSearchCV(forest, {'n_estimators': (1, 2), 'max_depth': (1, 2)})
     clf.fit(iris.data, iris.target)
 
 
+@pytest.mark.filterwarnings('ignore: The default of the `iid`')  # 0.22
 @pytest.mark.parametrize('name', FOREST_CLASSIFIERS)
 def test_gridsearch(name):
     # Check that base trees can be grid-searched.

diff --git a/sklearn/ensemble/tests/test_partial_dependence.py b/sklearn/ensemble/tests/test_partial_dependence.py
@@ -1,6 +1,7 @@
 """
 Testing for the partial dependence module.
 """
+import pytest
 
 import numpy as np
 from numpy.testing import assert_array_equal
@@ -103,6 +104,8 @@ def test_partial_dependecy_input():
     assert_raises(ValueError, partial_dependence, clf, [0], grid=grid)
 
 
+@pytest.mark.filterwarnings('ignore: Using or importing the ABCs from')
+# matplotlib Python3.7 warning
 @if_matplotlib
 def test_plot_partial_dependence():
     # Test partial dependence plot function.
@@ -135,6 +138,8 @@ def test_plot_partial_dependence():
     assert all(ax.has_data for ax in axs)
 
 
+@pytest.mark.filterwarnings('ignore: Using or importing the ABCs from')
+# matplotlib Python3.7 warning
 @if_matplotlib
 def test_plot_partial_dependence_input():
     # Test partial dependence plot function input checks.
@@ -170,6 +175,8 @@ def test_plot_partial_dependence_input():
                   clf, X, [{'foo': 'bar'}])
 
 
+@pytest.mark.filterwarnings('ignore: Using or importing the ABCs from')
+# matplotlib Python3.7 warning
 @if_matplotlib
 def test_plot_partial_dependence_multiclass():
     # Test partial dependence plot function on multi-class input.

diff --git a/sklearn/ensemble/tests/test_voting_classifier.py b/sklearn/ensemble/tests/test_voting_classifier.py
@@ -371,10 +371,12 @@ def test_set_estimator_none():
     X1 = np.array([[1], [2]])
     y1 = np.array([1, 2])
     eclf1 = VotingClassifier(estimators=[('rf', clf2), ('nb', clf3)],
-                             voting='soft', weights=[0, 0.5]).fit(X1, y1)
+                             voting='soft', weights=[0, 0.5],
+                             flatten_transform=False).fit(X1, y1)
 
     eclf2 = VotingClassifier(estimators=[('rf', clf2), ('nb', clf3)],
-                             voting='soft', weights=[1, 0.5])
+                             voting='soft', weights=[1, 0.5],
+                             flatten_transform=False)
     eclf2.set_params(rf=None).fit(X1, y1)
     assert_array_almost_equal(eclf1.transform(X1),
                               np.array([[[0.7, 0.3], [0.3, 0.7]],

diff --git a/sklearn/ensemble/tests/test_weight_boosting.py b/sklearn/ensemble/tests/test_weight_boosting.py
@@ -4,7 +4,7 @@
 import numpy as np
 
 from sklearn.utils.testing import assert_array_equal, assert_array_less
-from sklearn.utils.testing import assert_array_almost_equal
+from sklearn.utils.testing import assert_array_almost_equal, ignore_warnings
 from sklearn.utils.testing import assert_equal, assert_true, assert_greater
 from sklearn.utils.testing import assert_raises, assert_raises_regexp
 
@@ -196,6 +196,7 @@ def test_staged_predict():
     assert_array_almost_equal(score, staged_scores[-1])
 
 
+@pytest.mark.filterwarnings('ignore: The default of the `iid`')  # 0.22
 def test_gridsearch():
     # Check that base trees can be grid-searched.
     # AdaBoost classification

diff --git a/sklearn/ensemble/voting_classifier.py b/sklearn/ensemble/voting_classifier.py
@@ -91,7 +91,7 @@ class VotingClassifier(_BaseComposition, ClassifierMixin, TransformerMixin):
     >>> from sklearn.naive_bayes import GaussianNB
     >>> from sklearn.ensemble import RandomForestClassifier, VotingClassifier
     >>> clf1 = LogisticRegression(random_state=1)
-    >>> clf2 = RandomForestClassifier(random_state=1)
+    >>> clf2 = RandomForestClassifier(n_estimators=50, random_state=1)
     >>> clf3 = GaussianNB()
     >>> X = np.array([[-1, -1], [-2, -1], [-3, -2], [1, 1], [2, 1], [3, 2]])
     >>> y = np.array([1, 1, 1, 2, 2, 2])

diff --git a/sklearn/feature_extraction/tests/test_image.py b/sklearn/feature_extraction/tests/test_image.py
@@ -11,7 +11,8 @@
 from sklearn.feature_extraction.image import (
     img_to_graph, grid_to_graph, extract_patches_2d,
     reconstruct_from_patches_2d, PatchExtractor, extract_patches)
-from sklearn.utils.testing import assert_equal, assert_true, assert_raises
+from sklearn.utils.testing import (assert_equal, assert_true, assert_raises,
+                                   ignore_warnings)
 
 
 def test_img_to_graph():
@@ -55,6 +56,7 @@ def test_grid_to_graph():
     assert_true(A.dtype == np.float64)
 
 
+@ignore_warnings(category=DeprecationWarning)  # scipy deprecation inside face
 def test_connect_regions():
     try:
         face = sp.face(gray=True)
@@ -67,7 +69,7 @@ def test_connect_regions():
         graph = img_to_graph(face, mask)
         assert_equal(ndimage.label(mask)[1], connected_components(graph)[0])
 
-
+@ignore_warnings(category=DeprecationWarning)  # scipy deprecation inside face
 def test_connect_regions_with_grid():
     try:
         face = sp.face(gray=True)

diff --git a/sklearn/feature_extraction/tests/test_text.py b/sklearn/feature_extraction/tests/test_text.py
@@ -730,6 +730,7 @@ def test_vectorizer_inverse_transform(Vectorizer):
         assert_array_equal(np.sort(terms), np.sort(terms2))
 
 
+@pytest.mark.filterwarnings('ignore: The default of the `iid`')  # 0.22
 def test_count_vectorizer_pipeline_grid_selection():
     # raw documents
     data = JUNK_FOOD_DOCS + NOTJUNK_FOOD_DOCS
@@ -766,6 +767,7 @@ def test_count_vectorizer_pipeline_grid_selection():
     assert_equal(best_vectorizer.ngram_range, (1, 1))
 
 
+@pytest.mark.filterwarnings('ignore: The default of the `iid`')  # 0.22
 def test_vectorizer_pipeline_grid_selection():
     # raw documents
     data = JUNK_FOOD_DOCS + NOTJUNK_FOOD_DOCS

diff --git a/sklearn/feature_extraction/text.py b/sklearn/feature_extraction/text.py
@@ -14,7 +14,7 @@
 from __future__ import unicode_literals, division
 
 import array
-from collections import Mapping, defaultdict
+from collections import defaultdict
 import numbers
 from operator import itemgetter
 import re
@@ -32,6 +32,8 @@
 from .stop_words import ENGLISH_STOP_WORDS
 from ..utils.validation import check_is_fitted, check_array, FLOAT_DTYPES
 from ..utils.fixes import sp_version
+from ..utils.fixes import _Mapping as Mapping  # noqa
+
 
 __all__ = ['CountVectorizer',
            'ENGLISH_STOP_WORDS',

diff --git a/sklearn/feature_selection/base.py b/sklearn/feature_selection/base.py
@@ -72,7 +72,7 @@ def transform(self, X):
         X_r : array of shape [n_samples, n_selected_features]
             The input samples with only the selected features.
         """
-        X = check_array(X, accept_sparse='csr')
+        X = check_array(X, dtype=None, accept_sparse='csr')
         mask = self.get_support()
         if not mask.any():
             warn("No features were selected: either the data is"
@@ -111,7 +111,7 @@ def inverse_transform(self, X):
             return Xt
 
         support = self.get_support()
-        X = check_array(X)
+        X = check_array(X, dtype=None)
         if support.sum() != X.shape[1]:
             raise ValueError("X has a different shape than during fitting.")