[MRG+1] fixed OOB_Score bug for bagging classifiers. (scikit-learn#8936)

* fixed OOB_Score bug for bagging slassifiers.
See: scikit-learn#8933

* Added white space

* more white space fixing

* Adding test for oob_score validity

* removing pandas, replacing with numpy matrices

* fixing white space

* more white space fixing

* white space ...

* fixed labels to allow for strings

* white space

* simplifying test

* white space

* reformatting test

* white space

* pressed enter at end of file

* removing line at end of file

sklearn/ensemble/bagging.py

@@ -608,8 +608,7 @@ def _set_oob_score(self, X, y):
 
         oob_decision_function = (predictions /
                                  predictions.sum(axis=1)[:, np.newaxis])
-        oob_score = accuracy_score(y, classes_.take(np.argmax(predictions,
-                                                              axis=1)))
+        oob_score = accuracy_score(y, np.argmax(predictions, axis=1))
 
         self.oob_decision_function_ = oob_decision_function
         self.oob_score_ = oob_score

sklearn/ensemble/tests/test_bagging.py

@@ -723,3 +723,20 @@ def test_max_samples_consistency():
                                 max_features=0.5, random_state=1)
     bagging.fit(X, y)
     assert_equal(bagging._max_samples, max_samples)
+
+
+def test_set_oob_score_label_encoding():
+    # Make sure the oob_score doesn't change when the labels change
+    # See: https://github.com/scikit-learn/scikit-learn/issues/8933
+    randState = 5
+    X = [[-1], [0], [1]] * 5
+    Y1 = ['A', 'B', 'C'] * 5
+    Y2 = [-1, 0, 1] * 5
+    Y3 = [0, 1, 2] * 5
+    x1 = BaggingClassifier(oob_score=True,
+                           random_state=randState).fit(X, Y1).oob_score_
+    x2 = BaggingClassifier(oob_score=True,
+                           random_state=randState).fit(X, Y2).oob_score_
+    x3 = BaggingClassifier(oob_score=True,
+                           random_state=randState).fit(X, Y3).oob_score_
+    assert_equal([x1, x2], [x3, x3])