On Mon, Mar 5, 2012 at 3:15 PM, Jinhui Li <[email protected]> wrote: > The fit function of basedecisionregressor convert the X to dense format. > > In my case, there are 100M train samples, 30K features, most data are > zeros.
No one is working on sparse decision trees AFAIK. I'd suggest trying a dimensionality reduction approach (e.g. RandomizedPCA) and/or taking a random subsample (e.g 1M) of your data. Ciao Paolo ------------------------------------------------------------------------------ Try before you buy = See our experts in action! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-dev2 _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
