[Pkg-exppsy-maintainers] Continuous-label Classifier

[Yaroslav Halchenko]

well... we didn't come to tune up the tools for continuous labels yet
but they simply might work (I meant all kinds of custom classifiers in
mvpa.clf.classifier), but there is one basic classifier which can do
regression now (ie effectively do continuous label classification if I
get it right) -- SVMBase with appropriate parameter... just see help for
its __init__, it is pretty much just an adapter (to fit our
schema) around libsvm's python binding.

It would be great if you share how far you get with that one.

On Sat, 12 Jan 2008, Per B. Sederberg wrote:

> Hi Folks:

> Before I go coding up a ridge regression classifier for pymvpa (I'll
> likely do it anyway, I just would like to analyze some data faster
> than having to code it first), I figured I would ask if there is an
> existing classifier that can classify continuous labels.
> Specifically, I have TRs that do not have binary labels, but instead
> have a continuous label between 0 and 1, which I would like to predict
> in a cross-validated fashion.

> Thanks for any thoughts/tips,
> Per

> _______________________________________________
> Pkg-exppsy-maintainers mailing list
> Pkg-exppsy-maintainers at lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/pkg-exppsy-maintainers


-- 
Yaroslav Halchenko
Research Assistant, Psychology Department, Rutgers-Newark
Student  Ph.D. @ CS Dept. NJIT
Office: (973) 353-5440x263 | FWD: 82823 | Fax: (973) 353-1171
        101 Warren Str, Smith Hall, Rm 4-105, Newark NJ 07102
WWW:     http://www.linkedin.com/in/yarik