[pymvpa] Question about the searchlight example

[Michael Hanke]

On Mon, Mar 30, 2009 at 02:56:05PM +0200, Valentin Haenel wrote:
> Hi,
> 
> there is a line in the searchlioght example that i don't understand:
> 
> this line is fine, it will convert the one dimensional sl_map back into a 3
> dimensional volume.
> >     # map sensitivity map into original dataspace
> >     orig_sl_map = dataset.mapReverse(N.array(sl_map))
> 
> But what does this one do:?
> >     masked_orig_sl_map = N.ma.masked_array(orig_sl_map,
> >                                            mask=orig_sl_map == 0)
> 
> (Although maybe you just ought to point me in the direction of the masked array
> utils in numpy:)

This is purely to get transparency in the figure -- otherwise the error
overlay would totally obscure the image, while with masking we can
limit the visible pixels to those from within the brain.

Just visualization.

Michael

-- 
GPG key:  1024D/3144BE0F Michael Hanke
http://apsy.gse.uni-magdeburg.de/hanke
ICQ: 48230050