Setup

setpref('factorSpace', 'data_directory', '/path/to/data');
DATADIR = getpref('factorSpace', 'data_directory');

Data session files are structured as follows (n.b. * is a wildcard operator):

/DATADIR
   /Jeffy
      /20120302
          *simpleData_combined.mat
          kalmanInitParamsFA*.mat
      /20120303
      ...
   /Lincoln
   /Nelson
   ...

Processing new sessions

To process a list of session files for some dates, e.g., dts = {'20160101', '20160102'}, run the following:

1. Preprocess: cellfun(@io.saveDataByData, dts);
2. Fit behavioral asymptotes: Edit and run behav.asymptotesAll, making sure to add to the existing file 'data/asymptotes/bySession.mat'.
3. Fit IME: Run imefit.fitAll for your dts with opts = struct('doCv', false, 'doSave', true, 'fitPostLearnOnly', true). For each dtstr, you should verify there is now a file in fullfile(DATADIR, 'ime', [dtstr '.mat'])