API feature_selection's constructor params -> kwonly (scikit-learn#16867)

Author: adrinjalali

sklearn/feature_selection/_from_model.py

@@ -10,6 +10,7 @@
 
 from ..exceptions import NotFittedError
 from ..utils.metaestimators import if_delegate_has_method
+from ..utils.validation import _deprecate_positional_args
 
 
 def _get_feature_importances(estimator, norm_order=1):
@@ -157,7 +158,8 @@ class SelectFromModel(MetaEstimatorMixin, SelectorMixin, BaseEstimator):
            [-0.48],
            [ 1.48]])
     """
-    def __init__(self, estimator, threshold=None, prefit=False,
+    @_deprecate_positional_args
+    def __init__(self, estimator, *, threshold=None, prefit=False,
                  norm_order=1, max_features=None):
         self.estimator = estimator
         self.threshold = threshold

sklearn/feature_selection/_rfe.py

@@ -13,6 +13,7 @@
 from ..utils.metaestimators import if_delegate_has_method
 from ..utils.metaestimators import _safe_split
 from ..utils.validation import check_is_fitted
+from ..utils.validation import _deprecate_positional_args
 from ..base import BaseEstimator
 from ..base import MetaEstimatorMixin
 from ..base import clone
@@ -95,7 +96,7 @@ class RFE(SelectorMixin, MetaEstimatorMixin, BaseEstimator):
     >>> from sklearn.svm import SVR
     >>> X, y = make_friedman1(n_samples=50, n_features=10, random_state=0)
     >>> estimator = SVR(kernel="linear")
-    >>> selector = RFE(estimator, 5, step=1)
+    >>> selector = RFE(estimator, n_features_to_select=5, step=1)
     >>> selector = selector.fit(X, y)
     >>> selector.support_
     array([ True,  True,  True,  True,  True, False, False, False, False,
@@ -119,7 +120,8 @@ class RFE(SelectorMixin, MetaEstimatorMixin, BaseEstimator):
            for cancer classification using support vector machines",
            Mach. Learn., 46(1-3), 389--422, 2002.
     """
-    def __init__(self, estimator, n_features_to_select=None, step=1,
+    @_deprecate_positional_args
+    def __init__(self, estimator, *, n_features_to_select=None, step=1,
                  verbose=0):
         self.estimator = estimator
         self.n_features_to_select = n_features_to_select
@@ -464,7 +466,8 @@ class RFECV(RFE):
            for cancer classification using support vector machines",
            Mach. Learn., 46(1-3), 389--422, 2002.
     """
-    def __init__(self, estimator, step=1, min_features_to_select=1, cv=None,
+    @_deprecate_positional_args
+    def __init__(self, estimator, *, step=1, min_features_to_select=1, cv=None,
                  scoring=None, verbose=0, n_jobs=None):
         self.estimator = estimator
         self.step = step

sklearn/feature_selection/_univariate_selection.py

@@ -17,6 +17,7 @@
                      safe_mask)
 from ..utils.extmath import safe_sparse_dot, row_norms
 from ..utils.validation import check_is_fitted
+from ..utils.validation import _deprecate_positional_args
 from ._base import SelectorMixin
 
 
@@ -419,9 +420,9 @@ class SelectPercentile(_BaseFilter):
     SelectFwe: Select features based on family-wise error rate.
     GenericUnivariateSelect: Univariate feature selector with configurable mode.
     """
-
-    def __init__(self, score_func=f_classif, percentile=10):
-        super().__init__(score_func)
+    @_deprecate_positional_args
+    def __init__(self, score_func=f_classif, *, percentile=10):
+        super().__init__(score_func=score_func)
         self.percentile = percentile
 
     def _check_params(self, X, y):
@@ -503,9 +504,9 @@ class SelectKBest(_BaseFilter):
     SelectFwe: Select features based on family-wise error rate.
     GenericUnivariateSelect: Univariate feature selector with configurable mode.
     """
-
-    def __init__(self, score_func=f_classif, k=10):
-        super().__init__(score_func)
+    @_deprecate_positional_args
+    def __init__(self, score_func=f_classif, *, k=10):
+        super().__init__(score_func=score_func)
         self.k = k
 
     def _check_params(self, X, y):
@@ -582,9 +583,9 @@ class SelectFpr(_BaseFilter):
     SelectFwe: Select features based on family-wise error rate.
     GenericUnivariateSelect: Univariate feature selector with configurable mode.
     """
-
-    def __init__(self, score_func=f_classif, alpha=5e-2):
-        super().__init__(score_func)
+    @_deprecate_positional_args
+    def __init__(self, score_func=f_classif, *, alpha=5e-2):
+        super().__init__(score_func=score_func)
         self.alpha = alpha
 
     def _get_support_mask(self):
@@ -648,9 +649,9 @@ class SelectFdr(_BaseFilter):
     SelectFwe: Select features based on family-wise error rate.
     GenericUnivariateSelect: Univariate feature selector with configurable mode.
     """
-
-    def __init__(self, score_func=f_classif, alpha=5e-2):
-        super().__init__(score_func)
+    @_deprecate_positional_args
+    def __init__(self, score_func=f_classif, *, alpha=5e-2):
+        super().__init__(score_func=score_func)
         self.alpha = alpha
 
     def _get_support_mask(self):
@@ -711,9 +712,9 @@ class SelectFwe(_BaseFilter):
     SelectFdr: Select features based on an estimated false discovery rate.
     GenericUnivariateSelect: Univariate feature selector with configurable mode.
     """
-
-    def __init__(self, score_func=f_classif, alpha=5e-2):
-        super().__init__(score_func)
+    @_deprecate_positional_args
+    def __init__(self, score_func=f_classif, *, alpha=5e-2):
+        super().__init__(score_func=score_func)
         self.alpha = alpha
 
     def _get_support_mask(self):
@@ -761,7 +762,7 @@ class GenericUnivariateSelect(_BaseFilter):
     >>> X, y = load_breast_cancer(return_X_y=True)
     >>> X.shape
     (569, 30)
-    >>> transformer = GenericUnivariateSelect(chi2, 'k_best', param=20)
+    >>> transformer = GenericUnivariateSelect(chi2, mode='k_best', param=20)
     >>> X_new = transformer.fit_transform(X, y)
     >>> X_new.shape
     (569, 20)
@@ -786,8 +787,9 @@ class GenericUnivariateSelect(_BaseFilter):
                         'fdr': SelectFdr,
                         'fwe': SelectFwe}
 
-    def __init__(self, score_func=f_classif, mode='percentile', param=1e-5):
-        super().__init__(score_func)
+    @_deprecate_positional_args
+    def __init__(self, score_func=f_classif, *, mode='percentile', param=1e-5):
+        super().__init__(score_func=score_func)
         self.mode = mode
         self.param = param
 

sklearn/feature_selection/tests/test_feature_select.py

@@ -552,7 +552,7 @@ def test_nans():
     X = [[0, 1, 0], [0, -1, -1], [0, .5, .5]]
     y = [1, 0, 1]
 
-    for select in (SelectKBest(f_classif, 2),
+    for select in (SelectKBest(f_classif, k=2),
                    SelectPercentile(f_classif, percentile=67)):
         ignore_warnings(select.fit)(X, y)
         assert_array_equal(select.get_support(indices=True), np.array([1, 2]))