Should we rename optimize() to fit()?

[rhiever]

@rasbt, what do you think? We basically have a normal-ish sklearn interface here except for the optimize() call.

[rasbt]

In practice, I think it really doesn't matter ... However, it's probably not a bad idea down the road to ensure compatibility with scikit-learn functions. E.g., maybe you'd want to use it in a wrapper function like cross_val_score or so later on. Since it doesn't hurt, I'd recommend renaming it to fit :)