git pull https://github.com/korymath/gazevectorregression
Data files as *.mat files should be located in
/data
run_basic_exp will run on all *.mat files in the /data folder.
run_single_test provides a single experiment and you need to set train and test data files
Model is built in build_single_model.m
Can edit the way that the model is built by changing the mdl to other model builders from MATLAB.
modelBuildingIdeas.m has several ideas to try. I think that the best option will be a bagged regression, wishing there was a more automated trial set up in matlab.
One thing to do would be to output all the experiments to train and test directories so that we could rerun validations in Python autoML libraries, and faster c-modules to see if we could build up a solid single model.
all calibrations are sweep, eyes, free
the column switches between sweep and task
RXX - XX patient number
Calib(C|P) - cups or pasta, when this changes, task changes, always three for each assume new task mean adjustment
B - ignore (both eye and movement)
NN - 01, 02, 03 in sequence trials -- 1 and 2 are before then a whole bunch of trials, then 3 is the after trials calibration
_combined_segments - eye data,
TODO:
TESTING 1 on 2 and 1 on 3 compare