On Thu, Mar 22, 2012 at 08:42:03AM +0100, Andreas wrote: > > It is also my gut feeling that dividing the regularization term by > > n_samples make the optimal value *more* dependent on the dataset size > > rather that the opposite. That might be the reason why C is not scaled > > in the SVM literature. Off course I might be wrong as I have not done > > any kind of systematic cross-datasets analysis.
> I had the same feeling and I think we should really investigate this. > Volunteers? ;) If that turned out to be the case I would find it very interesting. I don't have any hard data on this. The hard data that we do have, is that if you compare SVM, Logistic l2 and Logistic l1, the SVM is less sensitive to the choice of regularisation factor, followed by the logistic l2, followed by the logistic l1. Gael ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
