def SVM_full_sequences_16itemsY(subject):
SVM_funcs.apply_SVM_filter_16_items_epochs_habituation(
subject,
times=[0.120, 0.190],
window=True,
sliding_window=True,
cleaned=False)
def SVM_full_sequences_16items1(subject):
# subject = config.subjects_list[0]
SVM_funcs.apply_SVM_filter_16_items_epochs(subject,
times=[0.130, 0.210],
window=True,
sliding_window=True,
cleaned=True)
def SVM_full_sequences_16items(subject):
# ----- We test on the 16 items sequences. We average the predictions of the decoders between 140 and 180 ms -----
SVM_funcs.apply_SVM_filter_16_items_epochs(subject,
times=[0.140, 0.180],
window=True,
sliding_window=True)
SVM_funcs.apply_SVM_filter_16_items_epochs_habituation(
subject, times=[0.140, 0.180], window=True, sliding_window=True)
def SVM_features_number_ofOpenedChunks(subject,
load_residuals_regression=False,
cleaned=True):
SVM_funcs.SVM_feature_decoding_wrapper(
subject,
'OpenedChunks',
SVM_dec=SVM_funcs.regression_decoder(),
balance_features=False,
distance=False,
load_residuals_regression=load_residuals_regression,
cross_val_func=None,
list_sequences=[3, 4, 5, 6, 7])
def SVM_features_stimID(subject,
load_residuals_regression=True,
cross_validation=None):
SVM_funcs.SVM_feature_decoding_wrapper(
subject,
'StimID',
load_residuals_regression=load_residuals_regression,
list_sequences=[3, 4, 5, 6, 7],
cross_val_func=cross_validation)
def SVM_features_number_ofOpenedChunks(subject,
load_residuals_regression=True):
if load_residuals_regression:
resid_suffix = 'resid_cv_'
else:
resid_suffix = 'full_data_'
score, distance, times = SVM_funcs.SVM_decode_feature(
subject,
'OpenedChunks',
SVM_dec=SVM_funcs.regression_decoder(),
load_residuals_regression=load_residuals_regression,
list_sequences=[3, 4, 5, 6, 7],
crop=[-0.1, 0.4],
cross_val_func=None,
balance_features=False,
distance=False)
save_name = config.SVM_path + subject + '/feature_decoding/' + resid_suffix + 'Number_Open_Chunks' + '_score_dict.npy'
np.save(save_name, {'score': score, 'times': times, 'distance': distance})
def SVM_features_repeatalter(subject,
load_residuals_regression=False,
cross_validation=None):
SVM_funcs.SVM_feature_decoding_wrapper(
subject,
'RepeatAlter',
load_residuals_regression=load_residuals_regression,
list_sequences=[3, 4, 5, 6, 7],
cross_val_func=cross_validation)
def SVM_features_withinchunk(subject,
load_residuals_regression=False,
cross_validation=None):
SVM_funcs.SVM_feature_decoding_wrapper(
subject,
'WithinChunkPosition',
load_residuals_regression=load_residuals_regression,
list_sequences=[4, 5, 6],
cross_val_func=cross_validation,
nvalues_feature=4)
def SVM_quad_ordpos(subject, cleaned=True):
SVM_funcs.SVM_feature_decoding_wrapper(
subject,
'WithinChunkPosition',
load_residuals_regression=False,
list_sequences=[4],
cross_val_func=None,
filter_from_metadata="StimPosition > 2 and StimPosition < 15",
nvalues_feature=4,
clean=cleaned)
def localize_standard_VS_deviant_code(subject,n_permutations = 2000,n_channels = 30,select_grad=False,cleaned=True):
# ----------- load the epochs ---------------
epochs = epoching_funcs.load_epochs_items(subject, cleaned=cleaned)
epochs.pick_types(meg=True)
# ----------- balance the position of the standard and the deviants -------
# 'local' - Just make sure we have the same amount of standards and deviants for a given position. This may end up with
# 1 standards/deviants for position 9 and 4 for the others.
epochs_balanced = epoching_funcs.balance_epochs_violation_positions(epochs,balance_param="local")
# ----------- do a sliding window to smooth the data -------
epochs_balanced = epoching_funcs.sliding_window(epochs_balanced)
# =============================================================================================
toi = 0.165
epochs_for_decoding = epochs_balanced.copy().crop(tmin=toi, tmax = toi)
training_inds, testing_inds = SVM_funcs.train_test_different_blocks(epochs_for_decoding, return_per_seq=False)
y_violornot = np.asarray(epochs_for_decoding.metadata['ViolationOrNot'].values)
labels_train = [y_violornot[training_inds[i]] for i in range(2)]
labels_test = [y_violornot[testing_inds[i]] for i in range(2)]
performance_loc = compute_sensor_weights_decoder(epochs_for_decoding,
SVM_funcs.SVM_decoder(),
training_inds,
labels_train,
testing_inds,
labels_test, None,
None, n_permutations,
n_channels,select_grad=select_grad)
suffix = ''
if select_grad:
suffix = 'only_grad'
save_path = config.result_path + '/localization/Standard_VS_Deviant/'
utils.create_folder(save_path)
save_path_subject = save_path + subject + '/'+suffix
utils.create_folder(save_path_subject)
np.save(save_path_subject + 'results'+str(n_permutations)+'_permut'+str(n_channels)+'_chans'+'_'+str(round(toi*1000))+'.npy', performance_loc)
def SVM_quad_ordpos(subject):
score, distance, times = SVM_funcs.SVM_decode_feature(
subject,
'WithinChunkPosition',
load_residuals_regression=True,
list_sequences=[4],
crop=[-0.1, 0.4],
cross_val_func=None,
filter_from_metadata="StimPosition > 2 and StimPosition < 15")
save_name = config.SVM_path + subject + '/feature_decoding/' + 'resid_' + 'WithinChunkPosition' + '_quads_score_dict.npy'
np.save(save_name, {'score': score, 'times': times, 'distance': distance})
score, distance, times = SVM_funcs.SVM_decode_feature(
subject,
'WithinChunkPosition',
load_residuals_regression=False,
list_sequences=[4],
crop=[-0.1, 0.4],
cross_val_func=None,
filter_from_metadata="StimPosition > 2 and StimPosition < 15")
save_name = config.SVM_path + subject + '/feature_decoding/' + 'full_data_' + 'WithinChunkPosition' + '_quads_score_dict.npy'
np.save(save_name, {'score': score, 'times': times, 'distance': distance})
def SVM_features_stimID_eeg(subject, load_residuals_regression=True):
if load_residuals_regression:
resid_suffix = 'resid_cv_'
else:
resid_suffix = 'full_data_'
score, distance, times = SVM_funcs.SVM_decode_feature(
subject,
'StimID',
load_residuals_regression=load_residuals_regression,
crop=[-0.1, 0.4],
cross_val_func=None,
meg=False)
save_name = config.SVM_path + subject + '/feature_decoding/' + resid_suffix + 'StimID' + '_EEGONLY_score_dict.npy'
np.save(save_name, {'score': score, 'times': times, 'distance': distance})
def SVM_features_chunkBeg(subject, load_residuals_regression=True):
if load_residuals_regression:
resid_suffix = 'resid_cv_'
else:
resid_suffix = 'full_data_'
score, distance, times = SVM_funcs.SVM_decode_feature(
subject,
'ChunkBeginning',
load_residuals_regression=load_residuals_regression,
list_sequences=[3, 4, 5, 6, 7],
crop=[-0.1, 0.4],
cross_val_func=None)
save_name = config.SVM_path + subject + '/feature_decoding/' + resid_suffix + 'ChunkBeg' + '_score_dict.npy'
np.save(save_name, {'score': score, 'times': times, 'distance': distance})
def SVM_features_withinchunk_train_quads_test_others(
subject, load_residuals_regression=True):
if load_residuals_regression:
resid_suffix = 'resid_cv_'
else:
resid_suffix = 'full_data_'
score, distance, times = SVM_funcs.SVM_decode_feature(
subject,
'WithinChunkPosition',
load_residuals_regression=load_residuals_regression,
crop=[-0.1, 0.4],
cross_val_func=SVM_funcs.train_quads_test_others,
balance_features=False,
filter_from_metadata="StimPosition > 2 and StimPosition < 15")
save_name = config.SVM_path + subject + '/feature_decoding/' + resid_suffix + 'WithinChunkPosition_train_Quads_test_others' + '_score_dict.npy'
np.save(save_name, {'score': score, 'times': times, 'distance': distance})
def SVM_GAT_all_sequences(subject):
SVM_funcs.GAT_SVM(subject,
load_residuals_regression=True,
sliding_window=True)
def GAT_SVM_separate_seq(subject):
SVM_funcs.GAT_SVM_trained_separate_sequences(
subject, load_residuals_regression=True, sliding_window=True)
def SVM_full_sequences_16items2(subject):
SVM_funcs.apply_SVM_filter_16_items_epochs(subject,
times=[0.210, 0.410],
window=True,
sliding_window=True,
cleaned=True)
def SVM_generate_different_sequences(subject):
# SVM_funcs.generate_SVM_separate_sequences(subject, load_residuals_regression=True,sliding_window=True)
SVM_funcs.GAT_SVM_trained_separate_sequences(
subject, load_residuals_regression=True, sliding_window=True)