search

TomMaullin / BLM

This repository contains all code for the python implementation of distributed OLS for locally stored data.
1 stars 3 forks source link

Add support for spatially varying predictors #63

Open nicholst opened 5 years ago

nicholst commented 5 years ago

(Low priority(!) item)

nicholst commented 5 years ago

How specify this...

X_files = ['/path/to/FirstSpatiallyVaryingRegressor.txt', 
           '/path/to/SecondSpatiallyVaryingRegressor.txt']

Each one of these files will have a list of N Nifti files inside.

And then user will need to know that contrasts will need that many more elements added at the end (relative to the non-spatially-varying design matrix X).

For now, no fancy masking options. I.e. if there are missing data in these spatially varying regressors, it is up to the user to modify the data masks to reflect this.