On Wed, Nov 09, 2011 at 11:25:17AM +0100, Andreas Müller wrote: > Here is an example: > https://gist.github.com/1351047 > It seems that at the moment, you have to use the whole training set to > generate the kernel at test time. > Not sure why, maybe for ease of use.
> Can anyone comment on that? I cannot comment on the whole thread, as I don't really use SVMs for my work, and certainly never precomputed kernels. That said, the questions raised are legitimate, and it would be great if the answers found their way in the docs, as a remark in the SVM docs (using the 'topic' directive) and as an example. I don't know what the best way to answer this usecase, but we probably need best-practices here. Gael ------------------------------------------------------------------------------ RSA(R) Conference 2012 Save $700 by Nov 18 Register now http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
