On Tue, Dec 6, 2011 at 10:37 AM, Peter Prettenhofer <peter.prettenho...@gmail.com> wrote: > 2011/12/6 James Bergstra <james.bergs...@gmail.com>: >> On Fri, Dec 2, 2011 at 12:54 PM, Peter Prettenhofer >> <peter.prettenho...@gmail.com> wrote: >>> [...] >>> >> >> How does the current tree implementation support boosting? I don't see >> anything in the code about weighted samples. >> >> - James > > You're right - we don't support sample weights at the moment but one > might use sampling with replacement to implement e.g. AdaBoost. > > Gradient boosting [1], on the other hand, does not need sample weights > but fits a series of regression trees on the residuals of their > predecessors. You can think of gradient boosting as a generalization > of boosting (forward stage-wise additive modelling) for arbitrary loss > functions (e.g. if you use exponential loss you recover AdaBoost) > > [1] http://en.wikipedia.org/wiki/Gradient_boosting
Thanks for pointing that out! That will be fine in my application. - James ------------------------------------------------------------------------------ Cloud Services Checklist: Pricing and Packaging Optimization This white paper is intended to serve as a reference, checklist and point of discussion for anyone considering optimizing the pricing and packaging model of a cloud services business. Read Now! http://www.accelacomm.com/jaw/sfnl/114/51491232/ _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
