How to use the memcnn.utils.log.setup function in memcnn

def main(data_dir, results_dir):
    # setup logging
    memcnn.utils.log.setup(True)

    # specify defaults for arguments
    use_cuda = torch.cuda.is_available()
    workers = 16
    experiments_file = os.path.join(os.path.dirname(__file__), 'config', 'experiments.json')
    start_fresh = False

    # parse arguments
    parser = argparse.ArgumentParser(description='Run memcnn experiments.')
    parser.add_argument('experiment_tags', type=str, nargs='+',
                        help='Experiment tags to run and combine from the experiment config file')
    parser.add_argument('--workers', dest='workers', type=int, default=workers,
                        help='Number of workers for data loading (Default: {})'.format(workers))
    parser.add_argument('--results-dir', dest='results_dir', type=str, default=results_dir,
                        help='Directory for storing results (Default: {})'.format(results_dir))
    parser.add_argument('--data-dir', dest='data_dir', type=str, default=data_dir,