[MRG + 2] Fixed parameter setting in SelectFromModel (scikit-learn#7764)

* Fixed cloning ``estimator`` again when calling fit a second time in SelectFromModel

* fix link in whatsnew

doc/whats_new.rst

@@ -141,6 +141,9 @@ Bug fixes
      functions were not accepting multi-label targets. :issue:`7676`
      by `Mohammed Affan`_
 
+   - Fixed setting parameters when calling ``fit`` multiple times on
+     :class:`feature_selection.SelectFromModel`. :issue:`7756` by `Andreas Müller`_
+
    - Fixes issue in ``partial_fit`` method of
      :class:`multiclass.OneVsRestClassifier` when number of classes used in
      ``partial_fit`` was less than the total number of classes in the

sklearn/feature_selection/from_model.py

@@ -232,8 +232,7 @@ def fit(self, X, y=None, **fit_params):
         if self.prefit:
             raise NotFittedError(
                 "Since 'prefit=True', call transform directly")
-        if not hasattr(self, "estimator_"):
-            self.estimator_ = clone(self.estimator)
+        self.estimator_ = clone(self.estimator)
         self.estimator_.fit(X, y, **fit_params)
         return self
 

sklearn/feature_selection/tests/test_from_model.py

@@ -2,6 +2,7 @@
 import scipy.sparse as sp
 
 from sklearn.utils.testing import assert_true
+from sklearn.utils.testing import assert_equal
 from sklearn.utils.testing import assert_less
 from sklearn.utils.testing import assert_greater
 from sklearn.utils.testing import assert_array_almost_equal
@@ -144,14 +145,13 @@ def test_partial_fit():
     assert_array_equal(X_transform, transformer.transform(data))
 
 
-def test_warm_start():
-    est = PassiveAggressiveClassifier(warm_start=True, random_state=0)
+def test_calling_fit_reinitializes():
+    est = LinearSVC(random_state=0)
     transformer = SelectFromModel(estimator=est)
     transformer.fit(data, y)
-    old_model = transformer.estimator_
+    transformer.set_params(estimator__C=100)
     transformer.fit(data, y)
-    new_model = transformer.estimator_
-    assert_true(old_model is new_model)
+    assert_equal(transformer.estimator_.C, 100)
 
 
 def test_prefit():