config.py

import argparse

parser = argparse.ArgumentParser()
parser.add_argument('--mode', type=str, default='synthesis',
                    help='predict or synthesis')
parser.add_argument('--batch_size', type=int, default=50,
                    help='minibatch size')
parser.add_argument('--chars', type=str, default='ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz ',
                    help='chars')
parser.add_argument('--num_epochs', type=int, default=200,
                    help='num of epochs')
parser.add_argument('--T', type=int, default=300,
                    help='RNN sequence length')
parser.add_argument('--points_per_char', type=int, default=25,
                    help='points per char (appr.)')
parser.add_argument('--rnn_state_size', type=int, default=400,
                    help='RNN hidden state size')
parser.add_argument('--num_layers', type=int, default=2,
                    help='num of RNN stack layers')
parser.add_argument('--M', type=int, default=20,
                    help='num of mixture bivariate gaussian')
parser.add_argument('--K', type=int, default=5,
                    help='num of mixture bivariate gaussian (for synthesis)')
parser.add_argument('--data_scale', type=float, default=20,
                    help='factor to scale raw data down by')
parser.add_argument('--learning_rate', type=float, default=0.001,
                    help='learning rate')
parser.add_argument('--b', type=float, default=3.0,
                    help='biased sampling')
parser.add_argument('--action', type=str, default='train',
                    help='mode')
# parser.add_argument('--decay_rate', type=float, default=0.95,
#                      help='decay rate for rmsprop')
# parser.add_argument('--keep_prob', type=float, default=0.8,
#                     help='dropout keep probability')
# parser.add_argument('--grad_clip', type=float, default=10.,
#                      help='clip gradients at this value')
args = parser.parse_args()