© 2013, 2014 Justis Grant Peters and Sagar Jauhari

Code repository for our project in Educational Data Mining class with Dr. Tiffany Barnes, CSC 591-021 Fall 2013 at NCSU. We are analyzing EEG data collected by Mostow and Chang from Project LISTEN.

$ python process_eeg.py -h
usage: process_eeg.py [-h] -i INDIR -o OUTDIR [--interpolate] [--kernelsvm]
                  	[--chartsforpaper]

Process EEG Data.

optional arguments:
  -h, --help        show this help message and exit
  -i INDIR          Directory containing raw EEG Files.
  -o OUTDIR         Path of processed data and reports. New directory with
                    timestamp will be created inside this folder.
  --interpolate     Interpolate 512Hz time stamps on 512Hz data having 1Hz
                    time stamps (raw data available from CMU)
  --kernelsvm       Perform Kernelized SVC on interpolated raw data
  --chartsforpaper  Create charts in single page PDF files.

The 512Hz data from CMU has timestamp information, but only at 1Hz resolution. To interpolate these:

python process_series_files.py data/raw/ preprocess/raw/

The Slicer class manages timeseries and extracted features as they relate to tasks from task.xls.

# instantiating task slicer
s = Slicer()

if len(sys.argv) > 2:
		print 'loading raw from list of csvfiles'
		s.load_series_from_csv('raw', sys.argv[1:])
else:
		print 'loading raw from pickle'
		s.load_series_from_pickle('raw', sys.argv[1])

# extract features from 'raw'
s.extract_filtered_signal()
s.extract_rolling_median()
s.extract_rolling_PSD()

# fetch task 1, with features
# Return type is a dict, with task-level features, plus the requested timeseries features
print s.get_by_task_id(1, features=['raw','raw_rolling_PSD_512', 'raw_rolling_median_128'])

sudo apt-get install ipython
git clone git@github.com:sagarjauhari/edm_eeg.git edm_eeg.git
cd edm_eeg.git
ipython notebook --pylab-inline

To run an R script, for example sketchpad.R:

cd dir-with-eeg-data/
R --vanilla < /path/to/sketchpad.R

It will produce a PDF called Rplots.pdf with a page for each chart produced by the script.

If you're using eeg_data_set2.zip_, you will want to fix the first line in 20101214171219.1.rawwave.csv to look like this:

%Time,Value

© 2013, 2014 Justis Grant Peters and Sagar Jauhari

This file is part of BCIpy.

BCIpy is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

BCIpy is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with BCIpy. If not, see http://www.gnu.org/licenses/.

• PyEEG - Licenced under GPL V3