I think working on speeding up trees would be a great topic for a gsoc. There is the idea of doing in-place resorting of X_argsorted and there are probably other ways we should investigate.
We could also do a "sparse matrix" gsoc adding sparse matrix support to all (?) estimators. Or maybe just to the trees? I won't be available for mentoring this year btw. Cheers, Andy ------------------------------------------------------------------------------ Free Next-Gen Firewall Hardware Offer Buy your Sophos next-gen firewall before the end March 2013 and get the hardware for free! Learn more. http://p.sf.net/sfu/sophos-d2d-feb _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
