Github user srowen commented on the pull request:
https://github.com/apache/spark/pull/160#issuecomment-38009810
Yeah actually this is a good partial step to distinguishing three different
things: giving an opaque score to a example/label, assigning a probability to
an example/label, and choosing the most likely label. The score/predict
distinction is not specific to binary classifiers. I suppose I'm saying: I may
propose a PR that would change this further, would that be OK? I think
`BinaryClassifierModel` might stay useful as at least a marker trait. Yes AUC
is useful.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
