On Mon, 15 Mar 2010, Jonas Kaplan wrote:
> The categories are indeed imbalanced across chunks, which given your
> previous paragraph is apparently something I need to keep a close eye
> on.
you might take advantage of having
nperlabel='equal', nrunspersplit=something_none_1
for your splitter... thus way it would assure that you have equal number
of samples of each category by subselecting... nrunspersplit would
subsample multiple times...
> Francisco, thanks for your clear explanation and those references look
> like they will be very useful.
I am taking my hat as well in a gesture of gratitude to Francisco for
the nice references ;)
--
.-.
=------------------------------ /v\ ----------------------------=
Keep in touch // \\ (yoh@|www.)onerussian.com
Yaroslav Halchenko /( )\ ICQ#: 60653192
Linux User ^^-^^ [175555]
_______________________________________________
Pkg-ExpPsy-PyMVPA mailing list
[email protected]
http://lists.alioth.debian.org/mailman/listinfo/pkg-exppsy-pymvpa
