On 10/22/2013 09:40 PM, Ralf Gunter wrote: > > However, I'm new to all of this and so have a related but potentially > dumb question. Don't worry about it ;)
I think there are two parts to the answer: GridSearchCV has a parameter "refit" which is True by default, which means that after taining, a new estimator will be fit on *all* data using the best parameters. The value that is reported as "best hyperparameters" is the mean over the scores of 5 different estimators. There is no single estimator yielding this score. You can use "best estimator" which will yield a similar score (it is probably one of the five that gives rise to the mean). Hth, Andy ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
