scikit-learn
diff --git a/‎sklearn/grid_search.py
Lines changed: 19 additions & 9 deletions b/‎sklearn/grid_search.py
Lines changed: 19 additions & 9 deletions
diff --git a/‎sklearn/tests/test_grid_search.py
Lines changed: 8 additions & 1 deletion b/‎sklearn/tests/test_grid_search.py
Lines changed: 8 additions & 1 deletion
@@ -340,7 +340,9 @@ def __init__(self, estimator, scoring=None, loss_func=None,
         self._check_estimator()
 
     def score(self, X, y=None):
-        """Returns the mean accuracy on the given test data and labels.
+        """Returns the score on the given test data and labels, if the search
+        estimator has been refit. The ``score`` function of the best estimator
+        is used, or the ``scoring`` parameter where unavailable.
 
         Parameters
         ----------
@@ -364,6 +366,22 @@ def score(self, X, y=None):
         y_predicted = self.predict(X)
         return self.scorer(y, y_predicted)
 
+    @property
+    def predict(self):
+        return self.best_estimator_.predict
+
+    @property
+    def predict_proba(self):
+        return self.best_estimator_.predict_proba
+
+    @property
+    def decision_function(self):
+        return self.best_estimator_.decision_function
+
+    @property
+    def transform(self):
+        return self.best_estimator_.transform
+
     def _check_estimator(self):
         """Check that estimator can be fitted and score can be computed."""
         if (not hasattr(self.estimator, 'fit') or
@@ -381,13 +399,6 @@ def _check_estimator(self):
                     "should have a 'score' method. The estimator %s "
                     "does not." % self.estimator)
 
-    def _set_methods(self):
-        """Create predict and  predict_proba if present in best estimator."""
-        if hasattr(self.best_estimator_, 'predict'):
-            self.predict = self.best_estimator_.predict
-        if hasattr(self.best_estimator_, 'predict_proba'):
-            self.predict_proba = self.best_estimator_.predict_proba
-
     def _fit(self, X, y, parameter_iterator, **params):
         "&q
8000
uot;"Actual fitting,  performing the search over parameters."""
         estimator = self.estimator
@@ -492,7 +503,6 @@ def _fit(self, X, y, parameter_iterator, **params):
             else:
                 best_estimator.fit(X, **self.fit_params)
             self.best_estimator_ = best_estimator
-            self._set_methods()
 
         # Store the computed scores
         CVScoreTuple = namedtuple('CVScoreTuple', ('parameters',
 
@@ -47,6 +47,10 @@ def fit(self, X, Y):
     def predict(self, T):
         return T.shape[0]
 
+    predict_proba = predict
+    decision_function = predict
+    transform = predict
+
     def score(self, X=None, Y=None):
         if self.foo_param > 1:
             score = 1.
@@ -133,8 +137,11 @@ def test_grid_search():
     for i, foo_i in enumerate([1, 2, 3]):
         assert_true(grid_search.cv_scores_[i][0]
                     == {'foo_param': foo_i})
-    # Smoke test the score:
+    # Smoke test the score etc:
     grid_search.score(X, y)
+    grid_search.predict_proba(X)
+    grid_search.decision_function(X)
+    grid_search.transform(X)
 
 
 def test_trivial_cv_scores():