def __init__(self, world, stage):
from nes_py.wrappers import JoypadSpace
import gym_super_mario_bros
from gym_super_mario_bros.actions import COMPLEX_MOVEMENT
env = gym_super_mario_bros.make(f"SuperMarioBros-{world}-{stage}-v0")
env = JoypadSpace(env, COMPLEX_MOVEMENT)
env = env_wrapper.SkipEnv(env, [])
env = env_wrapper.WarpFrame(env, width=84, height=84)
env = env_wrapper.MaxAndSkipEnv(env, 4)
env = env_wrapper.FrameStack(env, 4)
self._env = env
def _thunk():
env = gym.make(env_id)
env = env_wrapper.NoopResetEnv(env, noop_max=30)
env = env_wrapper.MaxAndSkipEnv(env, skip=4)
env.seed(rand_seed + rank)
env = env_wrapper.EpisodicLifeEnv(env)
if "FIRE" in env.unwrapped.get_action_meanings():
env = env_wrapper.FireResetEnv(env)
env = env_wrapper.WarpFrame(env)
env = env_wrapper.ClipRewardEnv(env)
return env
#----------------------------
n_stack = 4
env_id = args.env
save_dir = "./save_" + env_id
#Create multiple environments
#----------------------------
env = gym.make(env_id)
env = env_wrapper.NoopResetEnv(env, noop_max=30)
env = env_wrapper.MaxAndSkipEnv(env, skip=4)
env = env_wrapper.EpisodicLifeEnv(env)
if "FIRE" in env.unwrapped.get_action_meanings():
env = env_wrapper.FireResetEnv(env)
env = env_wrapper.WarpFrame(env)
env = env_wrapper.FrameStack(env, n_stack)
a_dim = env.action_space.n
img_height, img_width, c_dim = env.observation_space.shape
#Create the model
#----------------------------
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
policy = PolicyModel(sess, img_height, img_width, c_dim, a_dim)
#Start training
#----------------------------
sess.run(tf.global_variables_initializer())