BUG: avoid same indices in test and train

COSMIT: better names as suggested @schwarty

Address comments by @schwarty

Author: GaelVaroquaux

sklearn/cross_validation.py

@@ -1015,7 +1015,7 @@ def _iter_indices(self):
                                        )[0]
                 missing_idx = rng.permutation(missing_idx)
                 train.extend(missing_idx[:(self.n_train - len(train))])
-                test.extend(missing_idx[:(self.n_test - len(test))])
+                test.extend(missing_idx[-(self.n_test - len(test)):])
 
             train = rng.permutation(train)
             test = rng.permutation(test)

sklearn/tests/test_cross_validation.py

@@ -397,8 +397,8 @@ def test_stratified_shuffle_split_iter_no_indices():
     assert_array_equal(sorted(test_indices), np.where(test_mask)[0])
 
 
-def test_shuffle_split_even():
-    # Test the in StratifiedShuffleSplit, indices are drawn with a
+def test_stratified_shuffle_split_even():
+    # Test the StratifiedShuffleSplit, indices are drawn with a
     # equal chance
     n_folds = 5
     n_iter = 1000
@@ -431,6 +431,7 @@ def assert_counts_are_ok(idx_counts, p):
 
         assert_equal(len(train), splits.n_train)
         assert_equal(len(test), splits.n_test)
+        assert_equal(len(set(train).intersection(test)), 0)
 
         label_counts = np.unique(labels)
         assert_equal(splits.test_size, 1.0 / n_folds)