hi Andy, Lars is a forward feature selection technique. The parameter to cross-validate is the number of used features (like in OrthogonalMatchingPursuit btw) LarsCV should (in principle) estimate it.
Alex On Sat, Aug 18, 2012 at 9:27 PM, Andreas Mueller <[email protected]> wrote: > Hi everybody. > I was just trying to understand LarsCV. I am not so familiar with > the linear models so this might be a stupid question. > What parameter of Lars does LarsCV actually cross-validate? > It seems to me like the only important parameter of Lars is n_nonzero_coefs, > but I don't see how this can be obtained from LarsCV. > What am I missing? > > Best, > Andy > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Scikit-learn-general mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general > ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
