def main():
parser = argparse.ArgumentParser(
description='Eliminate the 1D subspace that correspond to BMI')
parser.add_argument('--in-data-dict-bin', type=str)
parser.add_argument('--in-feature-dim', type=int, default=20)
parser.add_argument('--out-data-dict-bin', type=str)
args = parser.parse_args()
in_dict_obj = load_object(args.in_data_dict_bin)
scan_name_list = list(in_dict_obj.keys())
data_X = np.zeros((len(scan_name_list), args.in_feature_dim), dtype=float)
data_Y = np.zeros((len(scan_name_list), ), dtype=float)
for idx_scan in range(len(scan_name_list)):
scan_name = scan_name_list[idx_scan]
data_X[idx_scan, :] = in_dict_obj[scan_name]['ImageData'][:]
data_Y[idx_scan] = in_dict_obj[scan_name]['bmi']
linear_reg_obj = EigenThoraxLinearRegression1D(data_X, data_Y)
linear_reg_obj.run_regression()
projected_data_X = linear_reg_obj.project_to_complement_space()
for idx_scan in range(len(scan_name_list)):
scan_name = scan_name_list[idx_scan]
in_dict_obj[scan_name]['ImageData'] = projected_data_X[idx_scan, :]
save_object(in_dict_obj, args.out_data_dict_bin)
def main():
parser = argparse.ArgumentParser(description='Load a saved pca object')
parser.add_argument('--low-dim-bin-path', type=str)
parser.add_argument('--save-bin-path', type=str)
parser.add_argument('--num-pca-component', type=int, default=10)
parser.add_argument('--dim-embedded', type=int, default=2)
args = parser.parse_args()
logger.info(f'Load low dim data from {args.low_dim_bin_path}')
low_dim_array = load_object(args.low_dim_bin_path)
data_matrix = np.zeros((len(low_dim_array), args.num_pca_component))
for sample_idx in range(len(low_dim_array)):
data_matrix[sample_idx, :] = low_dim_array[sample_idx]['low_dim'][:]
logger.info(f'Num of sample: {data_matrix.shape[0]}')
logger.info(f'Num of included PCs: {data_matrix.shape[1]}')
logger.info('Start tSNE')
# embedded_matrix = TSNE(perplexity=50, learning_rate=10000, n_components=args.dim_embedded).fit_transform(data_matrix)
embedded_matrix = TSNE(perplexity=50, n_iter=100000, n_components=args.dim_embedded).fit_transform(
data_matrix)
# embedded_matrix = TSNE(perplexity=50, learning_rate=10000, n_components=args.dim_embedded).fit_transform(
# data_matrix)
logger.info('Complete')
logger.info(f'Output shape: {embedded_matrix.shape}')
for sample_idx in range((len(low_dim_array))):
low_dim_array[sample_idx]['tsne_data'] = embedded_matrix[sample_idx, :]
# logger.info(low_dim_array[0])
logger.info(f'Save data to {args.save_bin_path}')
save_object(low_dim_array, args.save_bin_path)
def main():
parser = argparse.ArgumentParser(description='Load a saved pca object')
parser.add_argument('--in-pca-data-bin', type=str)
parser.add_argument('--label-file', type=str)
parser.add_argument('--out-data-dict-bin', type=str)
args = parser.parse_args()
low_dim_array = load_object(args.in_pca_data_bin)
label_obj = ClinicalDataReaderSPORE.create_spore_data_reader_xlsx(args.label_file)
data_dict = generate_data_dict(low_dim_array, label_obj)
logger.info(f'Save dict data object to {args.out_data_dict_bin}')
save_object(data_dict, args.out_data_dict_bin)
def run_dimension_reduction(self, save_bin_path):
pca_nii_3d = PCA_NII_3D(None, None, 1)
pca_nii_3d.load_pca(self._pca_bin_path)
image_feature_data_obj = load_object(self._data_bin_path)
projected_matrix = pca_nii_3d._get_pca().transform(image_feature_data_obj['data_matrix'])
out_data = {
'file_list': image_feature_data_obj['file_list'],
'projected_matrix': projected_matrix
}
save_object(out_data, save_bin_path)
def save_gaussian_model(self, out_bin):
save_object(self._gaussian_model, out_bin)
def save_data_dict_bin(self, bin_file_path):
save_object(self._data_dict, bin_file_path)
def run_dimension_reduction(self, save_bin_path):
self._low_dim_dict = self.run_parallel()
save_object(self._low_dim_dict, save_bin_path)
def save_pca_obj(self, file_path):
save_object(self._pca, file_path)
def save_bin(self, out_path):
save_object(self._in_dict_obj, out_path)