On 02/05/2013 07:09 PM, Lars Buitinck wrote: > 2013/2/5 Gael Varoquaux <[email protected]>: >> Do we need such genericity? Are there real practical gains to such >> modularity? My gut feeling would be to move forward with something >> simple, and make it performant statistical and computationaly for the >> common cases. > Perhaps not, but would it hurt to have both? > > 2013/2/5 Olivier Grisel <[email protected]>: >> Argl. This does not feel like linear model model at all. I would >> rather put the transformer in sklearn.kernel_approximation . > Indeed it doesn't :) +1 for kernel_approximations. > I feel it should be next to the kernel approximations but I don't like the module name :-/ I would call it feature extraction but that is already taken. And "feature learning" doesn't really fit if there is no learning...
------------------------------------------------------------------------------ Free Next-Gen Firewall Hardware Offer Buy your Sophos next-gen firewall before the end March 2013 and get the hardware for free! Learn more. http://p.sf.net/sfu/sophos-d2d-feb _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
