1b13fc95c046cb94c591c3cbbd5b5abed641bf27,examples/ppo_gym.py,,,#,6

Before Change

is_disc_action = len(env.action_space.shape) == 0
ActionTensor = LongTensor if is_disc_action else DoubleTensor

running_state = ZFilter((state_dim,), clip=5)
running_reward = ZFilter((1,), demean=False, clip=10)

define actor and critic
if is_disc_action:

After Change

                    help="maximal number of main iterations (default: 1000)")
parser.add_argument("--log-interval", type=int, default=1, metavar="N",
                    help="interval between training status logs (default: 10)")
parser.add_argument("--save-model-interval", type=int, default=0, metavar="N",
                    help="interval between saving model (default: 0, means don"t save)")
args = parser.parse_args()

env = gym.make(args.env_name)
env.seed(args.seed)

Instances