def get_bci_iii_3a(path, subject=1, session=1, task=1):
if task > 1:
print "This dataset has just one task!"
return None
if session > 1:
print "This dataset has only one session per subject!"
return None
description = 'BCI III Competition - Dataset IIIa'
ch_names = []
ch_names = ch_names + ['ch' + str(i + 1) for i in range(60)]
ch_names = ch_names + ['stim']
ch_types = []
ch_types = ch_types + ['eeg'] * 60
ch_types = ch_types + ['stim']
event_id = dict(left=1, right=2, feet=3, tongue=4)
filename = 'subject' + str(subject) + '.mat'
filepath = path + filename
mat = loadmat(filepath)
fs = mat['fs']
signal = mat['signal'].T.astype('float') # nchannels, nsamples
raw = conversion.create_mne_raw(signal, fs, ch_names, ch_types,
description)
return raw, event_id
def get_brain_invaders_p300(path, subject=1, session=1, task=1):
if task > 1:
print "This dataset has just one task!"
return None
if session > 1:
print "This dataset has just one session per subject!"
return None
description = 'P300 from Brain Invaders'
ch_names = []
ch_names = ch_names + ['ch' + str(i + 1) for i in range(32)]
ch_names = ch_names + ['stim']
event_id = dict(target=2, nontarget=1)
ch_types = []
ch_types = ch_types + ['eeg'] * 32
ch_types = ch_types + ['stim']
filename = 'subject' + "{0:0>2}".format(subject) + '.mat'
filepath = path + filename
mat = loadmat(filepath)
fs = mat['fs']
signal = mat['signal'].T.astype('float') # nchannels, nsamples
raw = conversion.create_mne_raw(signal, fs, ch_names, ch_types,
description)
return raw, event_id
def get_bci_iii_p300(path, subject=1, session=1, task=1):
if task > 1:
print "This dataset has just one task!"
return None
if session > 2:
print "This dataset has just two sessions per subject!"
return None
description = 'BCI III Competition - Dataset II (P300)'
ch_names = []
ch_names = ch_names + ['ch' + str(i + 1) for i in range(64)]
ch_names = ch_names + ['stim']
event_id = dict(target=2, nontarget=1)
ch_types = []
ch_types = ch_types + ['eeg'] * 64
ch_types = ch_types + ['stim']
filename = 'Subject' + str(subject) + '_part' + str(session) + '.mat'
filepath = path + filename
mat = loadmat(filepath)
fs = mat['fs']
signal = mat['signal'].T.astype('float') # nchannels, nsamples
raw = conversion.create_mne_raw(signal, fs, ch_names, ch_types,
description)
return raw, event_id
def get_bnci_erp_als(path, subject=1, session=1, task=1):
if task > 1:
print "This dataset has just one task!"
return None
if session > 1:
print "This dataset has just one session per subject!"
return None
description = 'P300 speller with ALS patients (from BNCI)'
ch_names = []
mat = loadmat(path + 'chnames.mat')
nchannels = mat['chnames'].shape[1]
ch_names = []
for n in range(nchannels):
ch_names.append(mat['chnames'][0][n][0])
ch_names += ['stim']
ch_types = []
ch_types = ch_types + ['eeg'] * nchannels
ch_types = ch_types + ['stim']
event_id = dict(target=2, nontarget=1)
filename = 'A0' + str(subject) + '.mat'
filepath = path + filename
mat = loadmat(filepath)
fs = mat['fs']
signal = mat['signal'].T.astype('float') # nchannels, nsamples
raw = conversion.create_mne_raw(signal, fs, ch_names, ch_types,
description)
return raw, event_id
def get_bnci_individual_imagery(path, subject=1, session=1, task=1):
if task > 1:
print "This dataset has only one task!"
return None
if session > 1:
print "This dataset has just two sessions per subject!"
return None
description = 'Individual motor imagery database (from BNCI)'
ch_names = []
mat = loadmat(path + 'chnames.mat')
nchannels = mat['chnames'].shape[1]
ch_names = []
for n in range(nchannels):
ch_names.append(mat['chnames'][0][n][0])
ch_names += ['stim']
ch_types = []
ch_types = ch_types + ['eeg'] * nchannels
ch_types = ch_types + ['stim']
event_id = dict(word=1, sub=2, nav=3, hand=4, feet=5)
subjects = ['A', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'L']
filename = str(subjects[subject]) + str(session) + '.mat'
filepath = path + filename
mat = loadmat(filepath)
fs = mat['fs']
signal = mat['signal'].T.astype('float') # nchannels, nsamples
raw = conversion.create_mne_raw(signal, fs, ch_names, ch_types,
description)
return raw, event_id
def get_bnci_two_class_motor_imagery(path, subject=1, session=1, task=1):
if task > 1:
print "This dataset has only one task!"
return None
if session > 1:
print "This dataset has only one session per subject!"
return None
description = 'Two class motor imagery database (from BNCI)'
ch_names = []
ch_names = ch_names + ['ch' + str(i + 1) for i in range(15)]
ch_names = ch_names + ['stim']
ch_types = []
ch_types = ch_types + ['eeg'] * 15
ch_types = ch_types + ['stim']
event_id = dict(right=1, feet=2)
# getting the training data for this subject
if subject < 10:
filename = 'S0' + str(subject) + 'T.mat'
else:
filename = 'S' + str(subject) + 'T.mat'
filepath = path + filename
mat = loadmat(filepath)
fs = mat['fs']
signal = mat['signal'].T.astype('float') # nchannels, nsamples
raw = conversion.create_mne_raw(signal, fs, ch_names, ch_types,
description)
return raw, event_id
def get_bci_iii_4a(path, subject=1, session=1, task=1):
if task > 1:
print "This dataset has just one task!"
return None
if session > 1:
print "This dataset has only one session per subject!"
return None
description = 'BCI III Competition - Dataset IVa'
ch_names = []
mat = loadmat(path + 'chnames.mat')
nchannels = mat['chnames'].shape[1]
ch_names = []
for n in range(nchannels):
ch_names.append(mat['chnames'][0][n][0])
ch_names += ['stim']
ch_types = []
ch_types = ch_types + ['eeg'] * nchannels
ch_types = ch_types + ['stim']
event_id = dict(hand=1, feet=2)
filename = 'subject' + str(subject) + '.mat'
filepath = path + filename
mat = loadmat(filepath)
fs = mat['fs']
signal = mat['signal'].T.astype('float') # nchannels, nsamples
raw = conversion.create_mne_raw(signal, fs, ch_names, ch_types,
description)
return raw, event_id
def get_bci_iv_2a(path, subject=1, session=1, task=1):
if task > 1:
print "This dataset has just one task!"
return None
if session > 2:
print "This dataset has just two sessions per subject!"
return None
description = 'BCI IV Competition - Dataset 2a'
ch_names = []
mat = loadmat(path + 'chnames.mat')
nchannels = mat['chnames'].shape[1]
ch_names = []
for n in range(nchannels):
ch_names.append(mat['chnames'][0][n][0])
ch_names = ch_names + ['eog' + str(i + 1) for i in range(3)]
ch_names += ['stim']
ch_types = []
ch_types = ch_types + ['eeg'] * 22
ch_types = ch_types + ['eog'] * 3
ch_types = ch_types + ['stim']
event_id = dict(left=1, right=2, feet=3, tongue=4)
suffix = ['T', 'E'][session - 1]
filename = 'A0' + str(subject) + suffix + '.mat'
filepath = path + filename
mat = loadmat(filepath)
fs = mat['fs']
signal = mat['signal'].T.astype('float') # nchannels, nsamples
raw = conversion.create_mne_raw(signal, fs, ch_names, ch_types,
description)
return raw, event_id