[pymvpa] saving searchlight results

[Yaroslav Halchenko]

> This did successfully save, however, when I opened the file in Afni the
> overlayed file was about 1/6 the size of my high res scan. Clearly, not the same
> space as my original dataset. How can I map the results into the same space as
> my original dataset? 

I guess you have meant that you've lost all metrics information (ie
voxel sizes, etc), lets see why that happened:

> orig_s1_map = dataset.mapReverse(N.array(s1_map))
here you got sensitivities (1D matrix) and mapped it back into 3D matrix
which has no information on voxel sizes, qform etc

> NiftiImage(orig_s1_map).save('searchlight_results.nii.gz')
here you created a new NiftiImage based on that 3D and stored it...

what is advised is to use .map2Nifti method of NiftiDataset, just look
into

doc/examples/searchlight.py:

so you would need just

dataset.map2Nifti(s1_map).save('...')

-- 
Yaroslav Halchenko
Research Assistant, Psychology Department, Rutgers-Newark
Student  Ph.D. @ CS Dept. NJIT
Office: (973) 353-1412 | FWD: 82823 | Fax: (973) 353-1171
        101 Warren Str, Smith Hall, Rm 4-105, Newark NJ 07102
WWW:     http://www.linkedin.com/in/yarik