ENH Scale the sum_of_confidences to (-0.5, 0.5)

raghavrv · amueller · commit a870b90bf5a0 · 2015-03-25T17:29:59.000-04:00
diff --git a/sklearn/multiclass.py b/sklearn/multiclass.py
@@ -569,9 +569,14 @@ def decision_function(self, X):
         if max_confidences == min_confidences:
             return votes
 
-        # Scale the sum_of_confidences to [-0.4, 0.4] and add it with votes
-        return votes + sum_of_confidences * \
-               (0.4 / max(abs(max_confidences), abs(min_confidences)))
+        # Scale the sum_of_confidences to (-0.5, 0.5) and add it with votes.
+        # The motivation is to use confidence levels as a way to break ties in
+        # the votes without switching any decision made based on a difference
+        # of 1 vote.
+        eps = np.finfo(sum_of_confidences.dtype).eps
+        max_abs_confidence = max(abs(max_confidences), abs(min_confidences))
+        scale = (0.5 - eps) / max_abs_confidence
+        return votes + sum_of_confidences * scale
 
 
 @deprecated("fit_ecoc is deprecated and will be removed in 0.18."
