Multi-class predict_proba

[amueller]

Hey Mathieu.
Is there a reason that predict_proba is not implemented for multi-class classification?
The multi-class log loss is the correct log loss so predict_proba should just be the exp'ed and normalized decision function, right? Or am I overlooking something?
I'm trying to get to some more or less calibrated logistic model.

[mblondel]

PR welcome :) I would also support the OVR log loss by normalizing the probabilities.

[amueller]

Will do / try ;) Thought I overlooked something.