'IN29', 'IN30', 'IN31', 'IN32', 'IN33', 'IN34', 'IN35', 'IN38', 'IN39',
'IN40', 'IN41', 'IN42', 'IN43', 'IN45', 'IN46'
]
num_subj = len(subj_list)
run_number_dict = {
'move': [3, 5],
'plan': [3, 4],
'color': [3, 4],
}
runs = run_number_dict[label]
# load mask file
mask_img = get_full_mask(data_dir)
for subj in subj_list:
print('starting run %s, %s label' % (subj, label))
for run in runs:
# load behavioral data
labels = get_behavior_data(behav_dir,
subj,
run,
label,
stratified_group=True)
# load fmri file
img = load_fmri_image(data_dir, subj, run, labels)
estimator = LinearDiscriminantAnalysis()
# target, index, run info initialize (in feedback condition)
info_df = pd.read_csv(
'/clmnlab/GL/fmri_data/MVPA/behaviors/targetID_fb_GL.tsv',
delimiter='\t')
chance_level = 1 / 4
info_df = info_df[info_df.run.isin(runs)
& (info_df.feedback == feedback_on)]
targets = np.array(info_df['target'])
indexes = list(info_df['trial'] - 1)
group = list(info_df['run'])
# load mask
mask_img = mtk.get_full_mask('/clmnlab/GL/fmri_data/masks/',
'full_mask.group.nii.gz')
for subj in subj_list:
# load t-value fMRI images
img_list = [
mtk.load_5d_fmri_image(data_dir + '%s/stats/tvals.%s.r%02d.nii' %
(subj, subj, run)) for run in runs
]
# check image size
for img in img_list:
assert img.shape == (96, 114, 96, 145)
# indexing image
imgs = nilearn.image.concat_imgs(img_list)
imgs = nilearn.image.index_img(imgs, indexes)
'IN35', 'IN38', 'IN39', 'IN40', 'IN41', 'IN42',
'IN43', 'IN45', 'IN46'
]
num_subj = len(subj_list)
run_number_dict = {
'move': [3, 5],
'plan': [3, 4],
'color': [3, 4],
}
runs = run_number_dict[label]
# load mask file
mask_img = mtk.get_full_mask(mask_dir)
for subj in subj_list:
print('starting run %s, %s label' % (subj, label))
labels = [
mtk.get_behavior_data(behav_dir, subj, run, label)
for run in runs
]
imgs = [
nilearn.image.index_img(
mtk.load_5d_fmri_image(data_dir + 'tvalsLSA.%s.r0%d.nii.gz' % (subj, run)),
label['order'] - 1)
for label, run in zip(labels, runs)
]
run_number_dict = {
'move': [3, 5],
'plan': [3, 4],
'color': [3, 4],
}
if estimator_name == 'svc':
estimator = LinearSVC()
elif estimator_name == 'scaled-svc':
estimator = Pipeline([('scale', StandardScaler()),
('svc', LinearSVC())])
roi_labels, roi_masks = mtk.load_rois(mask_dir + '*.nii.gz')
mask_img = mtk.get_full_mask('/clmnlab/IN/MVPA/LSS_betas/data/')
num_subj = len(subj_list)
runs = run_number_dict[label]
img_data = initial_images()
prefix = '%s_%s' % (label, estimator_name)
with open(stats_dir + '%s_roi_accuracies.tsv' % prefix, 'w') as file:
file.write(('%s\t' * (num_subj + 1) + '%s\n') %
('aal_label', 'mask_size', *subj_list))
for mask_name, mask in zip(roi_labels, roi_masks):
scores = perform_analysis()
'GA15', 'GA18', 'GA19', 'GA20', 'GA21', 'GA23', 'GA26', 'GA27', 'GA28',
'GB01', 'GB02', 'GB05', 'GB07', 'GB08', 'GB11', 'GB12', 'GB13', 'GB14',
'GB15', 'GB18', 'GB19', 'GB20', 'GB21', 'GB23', 'GB26', 'GB27', 'GB28'
]
num_subj = len(subj_list)
runs = [1, 2, 3, 4, 5, 6, 7]
group = [(i // 12) + 1 for i in range(96)]
# estimator initialize
if estimator_name == 'lda':
estimator = LinearDiscriminantAnalysis()
else:
raise ValueError('!! %s is unknown estimator name' % estimator_name)
mask_img = get_full_mask('/clmnlab/GA/MVPA/fullmask_GAGB/',
'full_mask.GAGB01to19.nii.gz')
for subj in subj_list:
print('starting run %s' % subj)
for run in runs:
img = load_5d_fmri_image(data_dir + 'tvals.%s.r%02d.nii.gz' %
(subj, run))
img = average_N_in_4d_image(img, n=10)
img = nilearn.image.index_img(img, range(1, 97))
X = img
y, chance_level = get_label_list_and_chance_level()
searchlight_img = run_searchlight(mask_img,
X,