def action_space(self):
if isinstance(self.inter.action_space, NoneSpace):
return NoneSpace()
assert isinstance(self.inter.action_space, spaces.Tuple)
sps = self.inter.action_space.spaces
if any([isinstance(sp, NoneSpace) for sp in sps]):
return NoneSpace()
numdim = sum([np.int(np.prod(sps[i].shape)) for i in range(len(sps))])
return spaces.Box(-1., 1., shape=(numdim, ))
def __init__(self,
env_id='Pomme1v1',
random_side=True,
agent_list=None,
rule_agents=[],
replay_dir=None):
super(pommerman_1v1, self).__init__(env_id,
random_side,
agent_list,
rule_agents,
replay_dir,
n_player=2)
ac_sp = self.env.action_space
self.action_space = spaces.Tuple([ac_sp] * 2)
self.observation_space = spaces.Tuple([NoneSpace(), NoneSpace()])
def observation_space(self):
if isinstance(self.inter.observation_space, NoneSpace):
return NoneSpace()
assert isinstance(self.inter.observation_space, spaces.Tuple)
low = np.concatenate([box.low for box in self.inter.observation_space.spaces])
high = np.concatenate([box.high for box in self.inter.observation_space.spaces])
return spaces.Box(low=low, high=high)
def __init__(self,
env_id='PommeTeam-v0',
random_side=True,
agent_list=None,
rule_agents=[],
replay_dir=None):
super(pommerman_2v2, self).__init__(env_id,
random_side,
agent_list,
rule_agents,
replay_dir,
n_player=4)
ac_sp = spaces.Tuple([spaces.Discrete(6)] * 2)
self.action_space = spaces.Tuple([ac_sp] * 2)
ob_sp = spaces.Tuple([NoneSpace(), NoneSpace()])
self.observation_space = spaces.Tuple([ob_sp] * 2)
def __init__(self,
team_size=2,
time_limit=45,
disable_walker_contacts=True,
team_num=2,
render_name="human"):
self.team_size = team_size
self.team_num = team_num
self.env = dm_soccer.load(self.team_size, time_limit,
disable_walker_contacts)
ac_sp_i = spaces.Box(low=-1.0, high=1.0, shape=(3, ), dtype=np.float32)
ac_sp = spaces.Tuple(
[spaces.Tuple(tuple([ac_sp_i] * self.team_size))] * self.team_num)
self.action_space = ac_sp
#print(self.action_space)
self.observation_space = spaces.Tuple([NoneSpace(), NoneSpace()])
self.timestep = None
odict_sp = {}
odict = self.env.observation_spec()
for key in odict[0]:
odict_sp[key] = spaces.Box(-np.inf,
np.inf,
shape=(np.int(
np.prod(odict[0][key].shape)), ))
self.observation_space = spaces.Tuple(
[spaces.Tuple([spaces.Dict(odict_sp)] * self.team_size)] *
self.team_num)
# render
render_mode_list = self.create_render_mode(render_name,
show=False,
return_pixel=True)
if render_mode_list is not None:
self.metadata['render.modes'] = list(render_mode_list.keys())
self.viewer = {key: None for key in render_mode_list.keys()}
else:
self.metadata['render.modes'] = []
self.render_mode_list = render_mode_list
# set seed
self._seed()
def observation_space(self):
inter_space = self.inter.observation_space
if isinstance(inter_space, NoneSpace):
return NoneSpace()
elif isinstance(inter_space, spaces.Box):
return spaces.Box(low=self.op(inter_space.low),
high=self.op(inter_space.high),
dtype=inter_space.dtype)
else:
assert isinstance(inter_space, spaces.Tuple)
sps = list(inter_space.spaces)
assert isinstance(sps[self.index], spaces.Box)
sps[self.index] = spaces.Box(low=self.op(sps[self.index].low),
high=self.op(sps[self.index].high),
dtype=sps[self.index].dtype)
return spaces.Tuple(tuple(sps))
def observation_space(self):
if isinstance(self.inter.observation_space, gym.spaces.Tuple):
return self.inter.observation_space[0]
else:
return NoneSpace()
def observation_space(self):
inter_space = self.inter.observation_space
if isinstance(inter_space, NoneSpace):
return NoneSpace()
else:
return stack_space(inter_space, self.k)
def observation_space(self):
if self.wrapper:
return self.wrapper.observation_space
else:
return NoneSpace()
def observation_space(self):
if isinstance(self.inter.observation_space, NoneSpace):
return NoneSpace()
else:
assert isinstance(self.inter.observation_space, spaces.Tuple)
return self.inter.observation_space.spaces[0]
def observation_space(self):
if isinstance(self.inter.observation_space, NoneSpace):
return NoneSpace()
assert isinstance(self.inter.observation_space, spaces.Dict)
return self.convert_OrderedDict(self.inter.observation_space.spaces)
def __init__(self, env, cheat_players=None):
super(AllObs, self).__init__(env)
self.observation_space = NoneSpace()
self.cheat_players = cheat_players
def __init__(self):
self._observation_space = NoneSpace()
self._action_space = NoneSpace()
self.steps = None