[pymvpa] How to do hyperalignment with the unit of trial

[Yaroslav Halchenko]

On Tue, 17 Nov 2015, 孔令军 wrote:


> Hi
> I had used  hyperalignment got great result for one experiment before.
> Now face another experiment , which has only two runs , eight trial of each run
> .
> The code do the hyperalignment with the unit of run defaultly , so the result
> is not ideality .
> Because of one half of the data are deserted .
> How can I modify the code so I could do hyperalignment with the unit of trial ,
> so I can get more data for training .
> Or if there are other ways for my new data ?

do you mean that you have only 8*2=16 samples altogether?

then unlikely hyperalignment would work -- it requires significantly
more samples to provide any reasonable transformation

-- 
Yaroslav O. Halchenko
Center for Open Neuroscience     http://centerforopenneuroscience.org
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834                       Fax: +1 (603) 646-1419
WWW:   http://www.linkedin.com/in/yarik