Are all datasets supposed to have mappers? My data set was loaded directly from a NIFTI file, using mvpa2.datasets.mri.fmri_dataset, and when I loaded it there was no mapper attached, though I attributed it target values and chunks. In other words, my original data set also does not have any mappers. Should I be manually attributing mappers to my data set (e.g., the searchlight call or the z-scoring call) and attach all those mappers as a Chain?
Thanks! Taku On Sat, Nov 2, 2013 at 11:02 AM, Yaroslav Halchenko <[email protected]>wrote: > do in ipython > > map2nifti? > > and check out its doc: > > Definition: map2nifti(dataset, data=None, imghdr=None, imgtype=None) > Docstring: > Maps data(sets) into the original dataspace and wraps it into an Image. > > Parameters > ---------- > dataset : Dataset > The mapper of this dataset is used to perform the reverse-mapping. > data : ndarray or Dataset, optional > The data to be wrapped into NiftiImage. If None (default), it > would wrap samples of the provided dataset. If it is a Dataset > instance -- takes its samples for mapping. > > > so if your res[1] doesn't have the mapper, just give as the first argument > the > original dataset which had it and which you gave to searchlight call > > On Fri, 01 Nov 2013, Taku Ito wrote: > > > Hi again, > > I'm having issues with saving searchlight maps back to NIFTI format... > > I've been using the mvpa2.datasets.mri.map2nifti module and I've tried > > both > > map2nifti(res[1])� > > #and� > > map2nifti(res[1]).save('s1.nii.gz') > > But both come back with the > > AttributeError: 'DatasetAttributesCollection' object has no attribute > > 'mapper' > > It seems that I have no mapper in my data set object - however, all > I've > > done on the original data set is take out some samples, z-score them, > and > > run an mvpa searchlight analysis with a NFoldPartitioner and Balancer > > (combined into a chainNode). �It seems that the dimensions of the > output > > data set are essentially the same, minus the fact that the number of > > samples was reduced from 360 to 150 or so. > > I have, been able to use the h5save() command for hdf5 files, but > would > > like to get it into NIFTI format so I can view the statistical maps in > > AFNI. � > > Please let me know if you have any insights! Your help is appreciated. > > Taku > -- > Yaroslav O. Halchenko, Ph.D. > http://neuro.debian.net http://www.pymvpa.org http://www.fail2ban.org > Senior Research Associate, Psychological and Brain Sciences Dept. > 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 > > _______________________________________________ > Pkg-ExpPsy-PyMVPA mailing list > [email protected] > http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa -- Takuya Ito Cognitive Control & Psychopathology Laboratory Washington University in St. Louis Cole Neuroscience Laboratory (http://www.mwcole.net/) Center for Molecular and Behavioral Neuroscience (CMBN) Rutgers-Newark University
_______________________________________________ Pkg-ExpPsy-PyMVPA mailing list [email protected] http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa
