scikit-learn · rth · Jul 5, 2020 · Jul 5, 2020 · Jul 5, 2020 · Jul 5, 2020
diff --git a/sklearn/calibration.py b/sklearn/calibration.py
@@ -32,8 +32,9 @@
 from .utils.validation import _deprecate_positional_args
 
 
-class CalibratedClassifierCV(BaseEstimator, ClassifierMixin,
-                             MetaEstimatorMixin):
+class CalibratedClassifierCV(ClassifierMixin,
+                             MetaEstimatorMixin,
+                             BaseEstimator):
     """Probability calibration with isotonic regression or logistic regression.
 
     This class uses cross-validation to both estimate the parameters of a

diff --git a/sklearn/discriminant_analysis.py b/sklearn/discriminant_analysis.py
@@ -128,8 +128,9 @@ def _class_cov(X, y, priors, shrinkage=None):
     return cov
 
 
-class LinearDiscriminantAnalysis(BaseEstimator, LinearClassifierMixin,
-                                 TransformerMixin):
+class LinearDiscriminantAnalysis(LinearClassifierMixin,
+                                 TransformerMixin,
+                                 BaseEstimator):
     """Linear Discriminant Analysis
 
     A classifier with a linear decision boundary, generated by fitting class

diff --git a/sklearn/ensemble/_hist_gradient_boosting/tests/test_gradient_boosting.py b/sklearn/ensemble/_hist_gradient_boosting/tests/test_gradient_boosting.py
@@ -392,7 +392,7 @@ def test_missing_values_minmax_imputation():
     # The implementation of MIA as an imputation transformer was suggested by
     # "Remark 3" in https://arxiv.org/abs/1902.06931
 
-    class MinMaxImputer(BaseEstimator, TransformerMixin):
+    class MinMaxImputer(TransformerMixin, BaseEstimator):
 
         def fit(self, X, y=None):
             mm = MinMaxScaler().fit(X)

diff --git a/sklearn/ensemble/tests/test_stacking.py b/sklearn/ensemble/tests/test_stacking.py
@@ -262,7 +262,7 @@ def test_stacking_classifier_drop_binary_prob():
     assert X_meta.shape[1] == 2
 
 
-class NoWeightRegressor(BaseEstimator, RegressorMixin):
+class NoWeightRegressor(RegressorMixin, BaseEstimator):
     def fit(self, X, y):
         self.reg = DummyRegressor()
         return self.reg.fit(X, y)
@@ -271,7 +271,7 @@ def predict(self, X):
         return np.ones(X.shape[0])
 
 
-class NoWeightClassifier(BaseEstimator, ClassifierMixin):
+class NoWeightClassifier(ClassifierMixin, BaseEstimator):
     def fit(self, X, y):
         self.clf = DummyClassifier(strategy='stratified')
         return self.clf.fit(X, y)

diff --git a/sklearn/linear_model/_logistic.py b/sklearn/linear_model/_logistic.py
@@ -1008,8 +1008,9 @@ def _log_reg_scoring_path(X, y, train, test, pos_class=None, Cs=10,
     return coefs, Cs, np.array(scores), n_iter
 
 
-class LogisticRegression(BaseEstimator, LinearClassifierMixin,
-                         SparseCoefMixin):
+class LogisticRegression(LinearClassifierMixin,
+                         SparseCoefMixin,
+                         BaseEstimator):
     """
     Logistic Regression (aka logit, MaxEnt) classifier.
 
@@ -1498,8 +1499,9 @@ def predict_log_proba(self, X):
         return np.log(self.predict_proba(X))
 
 
-class LogisticRegressionCV(LogisticRegression, BaseEstimator,
-                           LinearClassifierMixin):
+class LogisticRegressionCV(LogisticRegression,
+                           LinearClassifierMixin,
+                           BaseEstimator):
     """Logistic Regression CV (aka logit, MaxEnt) classifier.
 
     See glossary entry for :term:`cross-validation estimator`.

diff --git a/sklearn/metrics/_plot/tests/test_plot_precision_recall.py b/sklearn/metrics/_plot/tests/test_plot_precision_recall.py
@@ -63,7 +63,7 @@ def test_errors(pyplot):
 def test_error_bad_response(pyplot, response_method, msg):
     X, y = make_classification(n_classes=2, n_samples=50, random_state=0)
 
-    class MyClassifier(BaseEstimator, ClassifierMixin):
+    class MyClassifier(ClassifierMixin, BaseEstimator):
         def fit(self, X, y):
             self.fitted_ = True
             self.classes_ = [0, 1]

diff --git a/sklearn/model_selection/tests/test_search.py b/sklearn/model_selection/tests/test_search.py
@@ -1821,7 +1821,7 @@ def test_scalar_fit_param(SearchCV, param_search):
     # unofficially sanctioned tolerance for scalar values in fit_params
     # non-regression test for:
     # https://github.com/scikit-learn/scikit-learn/issues/15805
-    class TestEstimator(BaseEstimator, ClassifierMixin):
+    class TestEstimator(ClassifierMixin, BaseEstimator):
         def __init__(self, a=None):
             self.a = a
 

diff --git a/sklearn/multioutput.py b/sklearn/multioutput.py
@@ -60,7 +60,8 @@ def _partial_fit_estimator(estimator, X, y, classes=None, sample_weight=None,
     return estimator
 
 
-class _MultiOutputEstimator(BaseEstimator, MetaEstimatorMixin,
+class _MultiOutputEstimator(MetaEstimatorMixin,
+                            BaseEstimator,
                             metaclass=ABCMeta):
     @abstractmethod
     @_deprecate_positional_args

diff --git a/sklearn/svm/_classes.py b/sklearn/svm/_classes.py
@@ -10,8 +10,9 @@
 from ..utils.deprecation import deprecated
 
 
-class LinearSVC(BaseEstimator, LinearClassifierMixin,
-                SparseCoefMixin):
+class LinearSVC(LinearClassifierMixin,
+                SparseCoefMixin,
+                BaseEstimator):
     """Linear Support Vector Classification.
 
     Similar to SVC with parameter kernel='linear', but implemented in terms of