Andreas Mueller <amueller@...> writes:

> 
> Am 29.11.2012 23:45, schrieb Afik Cohen:
> >
> >
> > Hey Mathieu!
> >
> > Pretty much the only reason we wrap SGDClassifier in a OneVsRestClassifier 
is 
so
> > we can get predict_proba results on a per class basis. This fits our use 
case
> > better than getting prediction proabilities spread over all classes at 
once.
> > Hmm.. is there a way to do that without the OneVsRestClassifier wrapper?
> >
> Does that mean that you are doing multi-label classification, i.e. there
> is possibly more than one correct answer per example?
> Then I think there is no way to achieve this currently without 
> OneVsRestClassifier,
> though it would be easy enough to implement.
> 

No, we aren't doing multi-label classification, just multiclass. He was saying
we could just use SGDClassifier directly, which is true, but AFAIK there is no
way to get good prediction probability outputs on a per-class basis unless you
train binary classifiers by wrapping it in a OneVsRestClassifier() call... or 
is there?

Afik




------------------------------------------------------------------------------
Keep yourself connected to Go Parallel: 
BUILD Helping you discover the best ways to construct your parallel projects.
http://goparallel.sourceforge.net
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to