2012/3/12 Vikram Kamath <[email protected]>: > 1. Splits in CART are restricted to binary splits (a C4.5/C5.0 D-Tree is > m-ary)
All our learners work on numeric data, meaning categorical data must be split into binary features according to a one-of-K representation prior to handing it to a learner. So unless you also suggest an efficient way of representing categorical data, multiway splits aren't going to occur. (The easiest way from the implementation standpoint might be to let the user supply a boolean mask indicating which features are categorical.) -- Lars Buitinck Scientific programmer, ILPS University of Amsterdam ------------------------------------------------------------------------------ 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
