The data is *not* scaled. Though it makes svm happy and make everything
better. it makes me lose data and I don't need it in this case.
On 2 December 2012 11:55, Olivier Grisel <[email protected]> wrote:
> Maybe also check that the data is scaled to unit variance (or maybe
> take the sqrt or log of it if it's positive and spans several orders
> of magnitudes).
>
>
> ------------------------------------------------------------------------------
> Keep yourself connected to Go Parallel:
> DESIGN Expert tips on starting your parallel project right.
> http://goparallel.sourceforge.net/
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
------------------------------------------------------------------------------
Keep yourself connected to Go Parallel:
DESIGN Expert tips on starting your parallel project right.
http://goparallel.sourceforge.net/
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
