'SVMs do not directly provide probability estimates, these are calculated
using five-fold cross-validation, and thus performance can suffer'
http://scikit-learn.org/dev/modules/svm.html#svm
2012/4/6 Lars Buitinck <[email protected]>
> Op 5 april 2012 11:49 heeft <[email protected]> het volgende
> geschreven:
> > I was wondering whether it is possible to output probability estimates
> in sklearn.svm.LinearSVC.
> >
> > The underlying liblinear has the following option:
> > -b probability_estimates: whether to output probability estimates, 0 or
> 1 (default 0)
> >
> > But trying to activate it in scikit is not allowed:
> > TypeError: __init__() got an unexpected keyword argument 'probability'
>
> Liblinear probability estimates are supported in its logistic
> regression mode only, which in scikit-learn is available as
> sklearn.linear_model[.sparse].LogisticRegression:
>
> http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html
>
> --
> Lars Buitinck
> Scientific programmer, ILPS
> University of Amsterdam
>
>
> ------------------------------------------------------------------------------
> For Developers, A Lot Can Happen In A Second.
> Boundary is the first to Know...and Tell You.
> Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
> http://p.sf.net/sfu/Boundary-d2dvs2
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
------------------------------------------------------------------------------
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
