DOC fixes in BalancedBaggingClassifier

Author: glemaitre

imblearn/ensemble/classifier.py

@@ -136,7 +136,7 @@ class BalancedBaggingClassifier(BaggingClassifier):
     estimators_features_ : list of arrays
         The subset of drawn features for each base estimator.
 
-    classes_ : array of shape = [n_classes]
+    classes_ : array, shape (n_classes,)
         The classes labels.
 
     n_classes_ : int or list
@@ -145,7 +145,7 @@ class BalancedBaggingClassifier(BaggingClassifier):
     oob_score_ : float
         Score of the training dataset obtained using an out-of-bag estimate.
 
-    oob_decision_function_ : array of shape = [n_samples, n_classes]
+    oob_decision_function_ : ndarray, shape (n_samples, n_classes)
         Decision function computed with out-of-bag estimate on the training
         set. If n_estimators is small it might be possible that a data point
         was never left out during the bootstrap. In this case,