Additional metrics for binary classifier

[r0f1]

Hi,
Consider a binary classifier that outputs a probability between 0 and 1. In order to use ConfusionMatrix() I created a new variable that is true if this continuous probability is > 0.5.
Doing this, however, throws away some information. For example, I can no longer calculate the area under the ROC curve or calculate the average precision score.

Is there a class or function in pandas_ml similar to ConfusionMatrix.print_stats(), that calculates these (and possibly other) numbers?