base_feature, atlas, fwhm, node_size, weight_method) flag_nan_exists = False flag_incomplete = False flag_unexpected = False dataset = MLDataset() incomplete_processing[base_feature][weight_method] = dict() comb_nan_values[base_feature][weight_method] = dict() for ds_name in dataset_list: # out_dir = pjoin(proc_dir, 'graynet', '{}_{}_fwhm{}'.format(base_feature, atlas, fwhm)) out_dir = pjoin(base_dir, 'graynet_{]_{}'.format(expt_prefix, atlas)) meta_list = pjoin(base_dir, 'diagnosis_meta_data_{}.csv'.format(ds_name)) sample_ids, classes = run_workflow.get_metadata(meta_list) incomplete_processing[base_feature][weight_method][ds_name] = list( ) comb_nan_values[base_feature][weight_method][ds_name] = list() for sample in sample_ids: feat_path = pjoin(out_dir, sample, '{}_{}'.format(expt_id, file_ext)) if pexists(feat_path): graph = nx.read_graphml(feat_path) data = get_weights_order(graph, atlas_rois) idx_nan = np.logical_not(np.isfinite(data)) local_flag_nan_exists = np.count_nonzero(idx_nan) > 0 if local_flag_nan_exists: flag_nan_exists = True comb_nan_values[base_feature][weight_method][
proc_dir = pjoin(base_dir, dataset_name, 'processed') freesurfer_dir = pjoin(proc_dir, 'freesurfer') target_list_dir = pjoin(proc_dir, 'target_lists') vis_out_dir = pjoin(proc_dir, 'visualizations') if not pexists(vis_out_dir): os.mkdir(vis_out_dir) subject_id_list = pjoin(target_list_dir, 'graynet.compute.list') # meta_file = pjoin(target_list_dir, 'meta_4RTNI_and_PPMI.csv') meta_file = pjoin(target_list_dir, 'meta_{}.csv'.format(dataset_name)) for base_feature in features_freesurfer: id_list, classes = get_metadata(meta_file) class_set = list(set(classes.values())) class_set.sort() labels = {sub: class_set.index(cls) for sub, cls in classes.items()} out_path = pjoin( vis_out_dir, 'raw_features_{}_{}.MLDataset.pkl'.format(base_feature, '_'.join(class_set))) try: ds = MLDataset(filepath=out_path) except: traceback.print_exc() id_data = import_features(freesurfer_dir, id_list,