Ok, you have a point (or rather 2 points). +1 for shuffling. ----- Original message ----- > On Tue, Aug 20, 2013 at 5:23 PM, Joel Nothman > <[email protected]>wrote: > > > On Tue, Aug 20, 2013 at 5:12 PM, Gael Varoquaux < > > [email protected]> wrote: > > > > > > Currently, there is a reliance on the order of the data. > > > > > > Standard K-fold cross-validation relies on the order of the data, and > > > that might be a good thing, if it reflects the dependency structure > > > of the observations. > > > > > > > Seeing as it does a [fold_no::n_folds] sample, I don't think you're > > going to capture a dependency structure. > > > > (And that is preceded by an unstable argsort.)
------------------------------------------------------------------------------ Introducing Performance Central, a new site from SourceForge and AppDynamics. Performance Central is your source for news, insights, analysis and resources for efficient Application Performance Management. Visit us today! http://pubads.g.doubleclick.net/gampad/clk?id=48897511&iu=/4140/ostg.clktrk _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
