load_untouch_nii python equivalent?

[alormes...@gmail.com]

Hi, 

I'm trying to replicate a classifier created in MATLAB using Python. In the MATLAB script, the load_untouch_nii (to my knowledge) can be used to load in Nifti images; in the current analysis I'm looking at, we load in a mask. I was wondering what configurations load_untouch_nii uses; specifically, I want to know if it standardizes voxels or applies high/low pass filtering?

[Greg Detre]

I don't remember us writing a load_untouch_nii.m and I can't find it in the scripts.

If it's part of the NIH's afni_matlab brik-loading functionality, then you're better off contacting them.

On Sat, 25 Jan 2020 at 10:53, <alormes...@gmail.com> wrote:

Hi, 

I'm trying to replicate a classifier created in MATLAB using Python. In the MATLAB script, the load_untouch_nii (to my knowledge) can be used to load in Nifti images; in the current analysis I'm looking at, we load in a mask. I was wondering what configurations load_untouch_nii uses; specifically, I want to know if it standardizes voxels or applies high/low pass filtering?

--
You received this message because you are subscribed to the Google Groups "Princeton MVPA Toolbox for Matlab" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mvpa-toolbox...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mvpa-toolbox/e86d4779-bea9-49cb-9c36-0024d4440b9c%40googlegroups.com.

[Arlene]

Hi Greg, 

Thanks for the feedback. So I experimented with the Python script I currently have, and found that using a different algorithm for the optimization helped me get way closer results than I have in MATLAB. 

Since it appears that the L2 model that sklearn (Python) has different parameters than the one MVPA uses, I'm trying to figure out the cross validation parameters MVPA uses so I can match it. As such, I wanted to know if the L2 linear model from MVPA's 'train_L2_RLR' uses a specific intercept/constant that is added onto the decision function? 

To view this discussion on the web visit https://groups.google.com/d/msgid/mvpa-toolbox/CALmhVd6QPyPp5AH-Ys8F7PHH0cTg389rcM7gLqfC9EYr4Q8p2A%40mail.gmail.com.

[Greg Detre]

I'm a little confused. I don't remember a train_L2_RLR function... where are you seeing that?

TBH, I'm not maintaining the toolbox actively any more, so I probably can't help you if you can't tell from the code yourself.

To view this discussion on the web visit https://groups.google.com/d/msgid/mvpa-toolbox/CAFbENnkD5Hc9Z86xgrXdq-xfO21tO4SqByknG-s4M_5AfksdvA%40mail.gmail.com.