On Thu, Aug 01, 2013 at 07:11:22AM -0400, Josh Wasserstein wrote:
> Thanks Joel. That's my intuition , but I don't quite get why that would lead 
> to
> identical scores. Every CV split should yield different training and testing
> data, so even if the RBG stores trivial coefficients (and I still don't know
> what trivial would mean here), the support vectors would be different, and the
> performance would vary for the same exact kernel, wouldn't they?

You have a small number of test samples, thus you only have a small
number of error configurations that you can do: it is discrete.

G

------------------------------------------------------------------------------
Get your SQL database under version control now!
Version control is standard for application code, but databases havent 
caught up. So what steps can you take to put your SQL databases under 
version control? Why should you start doing it? Read more to find out.
http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to