from sklearn.cross_validation import KFold
from sklearn.metrics import confusion_matrix
import scot.xvschema
# The data set contains a continuous 45 channel EEG recording of a motor
# imagery experiment. The data was preprocessed to reduce eye movement
# artifacts and resampled to a sampling rate of 100 Hz. With a visual cue, the
# subject was instructed to perform either hand or foot motor imagery. The
# trigger time points of the cues are stored in 'triggers', and 'classes'
# contains the class labels. Duration of the motor imagery period was
# approximately six seconds.
from scot.datasets import fetch
midata = fetch("mi")[0]
raweeg = midata["eeg"]
triggers = midata["triggers"]
classes = midata["labels"]
fs = midata["fs"]
locs = midata["locations"]
# Set random seed for repeatable results
np.random.seed(42)
# Switch backend to scikit-learn
scot.backend.activate('sklearn')
import scot
from scot.varica import cspvarica
from scot.datatools import cut_segments
import scot.plotting as splot
# The data set contains a continuous 45 channel EEG recording of a motor
# imagery experiment. The data was preprocessed to reduce eye movement
# artifacts and resampled to a sampling rate of 100 Hz. With a visual cue, the
# subject was instructed to perform either hand or foot motor imagery. The
# trigger time points of the cues are stored in 'triggers', and 'classes'
# contains the class labels. Duration of the motor imagery period was
# approximately six seconds.
from scot.datasets import fetch
midata = fetch("mi")[0]
raweeg = midata["eeg"]
triggers = midata["triggers"]
classes = midata["labels"]
fs = midata["fs"]
locs = midata["locations"]
# Set random seed for repeatable results
np.random.seed(42)
# Prepare data
#
# Here we cut out segments from 3s to 4s after each trigger. This is right in
# the middle of the motor imagery period.
data = cut_segments(raweeg, triggers, 3 * fs, 4 * fs)