import matplotlib.pyplot as plt
from mne.decoding import GeneralizationAcrossTime
from mne.epochs import concatenate_epochs
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from jr.utils import OnlineReport
from jr.gat import force_predict, scorer_auc
from base import eeglab2mne
report = OnlineReport()
##############################################################################
# Load data
data_path = '/media/jrking/harddrive/Simon'
all_scores = dict(HL=list(), EU=list(), PR=list())
for subject in range(1, 33):
file_name = op.join(data_path, 's%i.mat' % subject)
#######################################################################
# Preprocessing to speed it up
epochs = eeglab2mne(file_name)
np.random.RandomState(1)
order = np.arange(len(epochs))
np.random.shuffle(order)
"""
Plot decoding results
"""
import numpy as np
import itertools
import os.path as op
import pickle
import matplotlib.pyplot as plt
from jr.utils import OnlineReport
from jr.plot import pretty_gat, pretty_decod, pretty_slices
data_path = '/media/jrking/harddrive/Simon'
report = OnlineReport()
# gather subjects scores
contrasts = ['HL', 'EU', 'PR']
all_scores = dict(HL=list(), EU=list(), PR=list(),
EU_H=list(), PR_H=list(),
EU_L=list(), PR_L=list())
for subject, contrast in itertools.product(range(1, 33), contrasts):
fname = op.join(data_path, 's%i_%s_scores.npy' % (subject, contrast))
scores = np.load(fname)
all_scores[contrast].append(scores)
# gather subscores
for subject, HL, contrast in itertools.product(
range(1, 33), [0, 1], ['EU', 'PR']):
subcontrast = '%s_%s' % (contrast, 'HL'[HL])
fname = op.join(data_path, 's%i_%s_scores.npy' % (subject, subcontrast))
scores = np.load(fname)
all_scores[subcontrast].append(scores)
