def __init__(self, env_id, n_agents):
env_path = UnityVecEnv.GetFilePath(env_id, n_agents=n_agents)
print("**** ", env_path)
env = UnityEnv(env_path, multiagent=True)
self.env = env
env.num_envs = env.number_agents
VecEnv.__init__(self, env.num_envs, env.observation_space,
env.action_space)
obs_space = env.observation_space
# self.keys, shapes, dtypes = obs_space_info(obs_space)
# self.buf_obs = { k: np.zeros((self.num_envs,) + tuple(shapes[k]), dtype=dtypes[k]) for k in self.keys }
# self.buf_dones = np.zeros((self.num_envs,), dtype=np.bool)
# self.buf_rews = np.zeros((self.num_envs,), dtype=np.float32)
self.buf_infos = [{} for _ in range(self.num_envs)]
# Fake Monitor
self.tstart = time.time()
self.results_writer = ResultsWriter("filename",
header={
"t_start": time.time(),
'env_id': env.spec
and env.spec.id
},
extra_keys=() + ())
self.reset_keywords = ()
self.info_keywords = ()
self.allow_early_resets = True
self.rewards = None
self.needs_reset = True
self.episode_rewards = []
self.episode_lengths = []
self.episode_times = []
self.total_steps = 0
self.current_reset_info = {
} # extra info about the current episode, that was passed in during reset()
def __init__(self, env_id):
print ("**** ", env_id, platform.system())
# env = UnityEnv(env_id, multiagent=True)
env_id = "hopper"
# env_id = "walker"
if platform.system() == 'Windows':
env_path = os.path.join('envs', env_id+'-x16', 'Unity Environment.exe')
elif platform.system() == 'Darwin': # MacOS
env_path = os.path.join('envs', env_id+'-x16')
elif platform.system() == 'Linux':
env_path = os.path.join('envs', env_id+'-x16')
print ("**** Override", env_path, env_id)
env = UnityEnv(env_path, multiagent=True)
self.env = env
env.num_envs = env.number_agents
VecEnv.__init__(self, env.num_envs, env.observation_space, env.action_space)
# obs_space = env.observation_space
# spec = env.spec
# self.keys, shapes, dtypes = obs_space_info(obs_space)
# self.buf_obs = { k: np.zeros((self.num_envs,) + tuple(shapes[k]), dtype=dtypes[k]) for k in self.keys }
# self.buf_dones = np.zeros((self.num_envs,), dtype=np.bool)
# self.buf_rews = np.zeros((self.num_envs,), dtype=np.float32)
self.buf_infos = [{} for _ in range(self.num_envs)]
# Fake Monitor
self.tstart = time.time()
self.results_writer = ResultsWriter(
"filename",
header={"t_start": time.time(), 'env_id' : env.spec and env.spec.id},
extra_keys=() + ()
)
self.reset_keywords = ()
self.info_keywords = ()
self.allow_early_resets = True
self.rewards = None
self.needs_reset = True
self.episode_rewards = []
self.episode_lengths = []
self.episode_times = []
self.total_steps = 0
self.current_reset_info = {} # extra info about the current episode, that was passed in during reset()
