>> Multinomial logistic regression with a tweaked coordinate descent for >> the case n_samples << n_features would be a great GSoC proposal to add >> on the wikipage. > > I'd go for a few other losses in addition, including square loss (i.e. > improving our Lasso/ElasticNet). Hey, it's a GSOC, it should be > ambitious. > >> Any volunteer to mentor it (I am thinking Gael or >> Alexandre mostly, Lords of the small n_samples Realm). > > I guess Alex and I can mentor.
FYI I've added this to the wiki page Alex ------------------------------------------------------------------------------ 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
