mne-tools
diff --git a/‎mne/decoding/search_light.py
Lines changed: 15 additions & 21 deletions b/‎mne/decoding/search_light.py
Lines changed: 15 additions & 21 deletions
diff --git a/‎mne/decoding/tests/test_search_light.py
Lines changed: 3 additions & 1 deletion b/‎mne/decoding/tests/test_search_light.py
Lines changed: 3 additions & 1 deletion
@@ -34,23 +34,11 @@ def __repr__(self):
         return repr_str + '>'
 
     def __init__(self, base_estimator, scoring=None, n_jobs=1):
-        from sklearn.metrics import make_scorer, get_scorer
-
         _check_estimator(base_estimator)
         self.base_estimator = base_estimator
         self.n_jobs = n_jobs
         self.scoring = scoring
 
-        # If scoring is None (default), the predictions are internally
-        # generated by estimator.score(). Else, we must first get the
-        # predictions based on the scorer.
-        if not isinstance(self.scoring, str):
-            self.scoring = (make_scorer(self.scoring) if self.scoring is
-                            not None else self.scoring)
-
-        elif self.scoring is not None:
-            self.scoring = get_scorer(self.scoring)
-
         if not isinstance(self.n_jobs, int):
             raise ValueError('n_jobs must be int, got %s' % n_jobs)
 
@@ -239,18 +227,29 @@ 
10000
def score(self, X, y):
         score : array, shape (n_samples, n_estimators)
             Score for each estimator / data slice couple.
         """
+        from sklearn.metrics import make_scorer, get_scorer
         self._check_Xy(X)
         if X.shape[-1] != len(self.estimators_):
             raise ValueError('The number of estimators does not match '
                              'X.shape[-1]')
 
+        # If scoring is None (default), the predictions are internally
+        # generated by estimator.score(). Else, we must first get the
+        # predictions based on the scorer.
+        if not isinstance(self.scoring, str):
+            scoring_ = (make_scorer(self.scoring) if self.scoring is
+                        not None else self.scoring)
+
+        elif self.scoring is not None:
+            scoring_ = get_scorer(self.scoring)
+
         # For predictions/transforms the parallelization is across the data and
         # not across the estimators to avoid memory load.
         parallel, p_func, n_jobs = parallel_func(_sl_score, self.n_jobs)
         n_jobs = min(n_jobs, X.shape[-1])
         X_splits = np.array_split(X, n_jobs, axis=-1)
         est_splits = np.array_split(self.estimators_, n_jobs)
-        score = parallel(p_func(est, self.scoring, X, y)
+        score = parallel(p_func(est, scoring_, X, y)
                          for (est, x) in zip(est_splits, X_splits))
 
         if n_jobs > 1:
@@ -400,18 +399,13 @@ class GeneralizationLight(SearchLight):
     ----------
     base_estimator : object
         The base estimator to iteratively fit on a subset of the dataset.
+    scoring : callable, string, defaults to None
+        Score function (or loss function) with signature
+        score_func(y, y_pred, **kwargs).
     n_jobs : int, optional (default=1)
         The number of jobs to run in parallel for both `fit` and `predict`.
         If -1, then the number of jobs is set to the number of cores.
     """
-    def __init__(self, base_estimator, n_jobs=1):
-        _check_estimator(base_estimator)
-        self.base_estimator = base_estimator
-        self.n_jobs = n_jobs
-
-        if not isinstance(self.n_jobs, int):
-            raise ValueError('n_jobs must be int, got %s' % n_jobs)
-
     def __repr__(self):
         repr_str = super(GeneralizationLight, self).__repr__()
         if hasattr(self, 'estimators_'):
 
@@ -72,7 +72,9 @@ def test_searchlight():
     sl2.fit(X, y)
     assert_array_equal(score1, sl2.score(X, y))
 
-    assert_raises(ValueError, SearchLight, LogisticRegression(), scoring='foo')
+    sl = SearchLight(LogisticRegression(), scoring='foo')
+    sl.fit(X, y)
+    assert_raises(ValueError, sl.score, X, y)
 
     sl = SearchLight(LogisticRegression())
     assert_equal(sl.scoring, None)