On 03/14/2012 08:58 PM, Olivier Grisel wrote: > Le 14 mars 2012 08:57, Satrajit Ghosh<[email protected]> a écrit : > >>> IMO We should record the fit durations and return the fastest among >>> the tied models. This is model agnostic and favoring fast convergence >>> is a nice feature for the user. >>> >> >> doesn't too many machine and code dependent parameters come into play that >> have nothing to do with the simplicity of the model? >> > I am not asserting that faster models for a given accuracy level will > always be the most regularized once (I am not sure this is even true, > I am sure we can find counter examples). I am just asserting that > faster parameter sets are a good way to arbitrarily break ties on > predictive accuracy as the user will prefer using parameters that lead > to faster models. > > Counterexample: KNN. higher k = more regularization.
Though not really a counterexample on a claim any one made, just came into my head ;) ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
