[pymvpa] Help needed untangling searchlight accuracy bias

[Mike E. Klein]

Ha! Yes, I'm very much beholden to the developers and community: I'm a
cog neuro person who admittedly lacks some fundamental skills here
(though trying!).

I wasn't getting that error… at the time of zscoring I hadn't yet
removed category #3, so it was seeing -exactly- 3 samples per chunk.

On Thu, Oct 20, 2011 at 3:51 PM, Yaroslav Halchenko
<debian at onerussian.com> wrote:
>> Ah! I think I see what it is then… I'm likely feeding something like
>> -1s and 1s (only!) into the searchlight!
>
> that is right ;)
>
> weren't you getting a warning of
>
>            warning("Z-scoring chunk-wise having a chunk with less than three "
>                    "samples will set features in these samples to either zero "
>                    "(with 1 sample in a chunk) "
>                    "or -1/+1 (with 2 samples in a chunk).")
>
> this kind? ;)
>
>> I'll try out your suggestions and reply shortly (just stepped away
>> from the computer)… thanks again!
>
> no problem -- glad to be of help -- we could do even more
> consulting for co-authorship ;)
>
> --
> =------------------------------------------------------------------=
> Keep in touch                                     www.onerussian.com
> Yaroslav Halchenko                 www.ohloh.net/accounts/yarikoptic
>
> _______________________________________________
> Pkg-ExpPsy-PyMVPA mailing list
> Pkg-ExpPsy-PyMVPA at lists.alioth.debian.org
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa