[MRG+1] Ridgecv normalize (scikit-learn#9302)

agramfort · jnothman · commit a6753f3ed38d · 2017-07-09T21:21:45.000+10:00
* FIX : normalize was not passed to grid search in RidgeCV

* update what's new
diff --git a/doc/whats_new.rst b/doc/whats_new.rst
@@ -471,10 +471,13 @@ Bug fixes
      by :user:`Andre Ambrosio Boechat <boechat107>`, :user:`Utkarsh Upadhyay
      <musically-ut>`, and `Joel Nothman`_.
 
-
    - Add ``data_home`` parameter to
      :func:`sklearn.datasets.fetch_kddcup99` by `Loic Esteve`_.
 
+   - Fix inconsistent results between :class:`linear_model.RidgeCV`
+     and :class:`linear_model.Ridge` when using ``normalize=True``
+     by `Alexandre Gramfort`_.
+
 API changes summary
 -------------------
 
diff --git a/sklearn/linear_model/ridge.py b/sklearn/linear_model/ridge.py
@@ -1119,7 +1119,8 @@ def fit(self, X, y, sample_weight=None):
                 raise ValueError("cv!=None and store_cv_values=True "
                                  " are incompatible")
             parameters = {'alpha': self.alphas}
-            gs = GridSearchCV(Ridge(fit_intercept=self.fit_intercept),
+            gs = GridSearchCV(Ridge(fit_intercept=self.fit_intercept,
+                                    normalize=self.normalize),
                               parameters, cv=self.cv, scoring=self.scoring)
             gs.fit(X, y, sample_weight=sample_weight)
             estimator = gs.best_estimator_
diff --git a/sklearn/linear_model/tests/test_ridge.py b/sklearn/linear_model/tests/test_ridge.py
@@ -383,6 +383,16 @@ def _test_ridge_loo(filter_):
     return ret
 
 
+def _test_ridge_cv_normalize(filter_):
+    ridge_cv = RidgeCV(normalize=True, cv=3)
+    ridge_cv.fit(filter_(10. * X_diabetes), y_diabetes)
+
+    gs = GridSearchCV(Ridge(normalize=True), cv=3,
+                      param_grid={'alpha': ridge_cv.alphas})
+    gs.fit(filter_(10. * X_diabetes), y_diabetes)
+    assert_equal(gs.best_estimator_.alpha, ridge_cv.alpha_)
+
+
 def _test_ridge_cv(filter_):
     ridge_cv = RidgeCV()
     ridge_cv.fit(filter_(X_diabetes), y_diabetes)
@@ -462,6 +472,7 @@ def check_dense_sparse(test_func):
 def test_dense_sparse():
     for test_func in (_test_ridge_loo,
                       _test_ridge_cv,
+                      _test_ridge_cv_normalize,
                       _test_ridge_diabetes,
                       _test_multi_ridge_diabetes,
                       _test_ridge_classifiers,
