def prepare_for_trainer(
cls,
env_spec: EnvironmentSpec,
config: Dict
) -> "AlgorithmSpec":
config_ = config.copy()
agents_config = config_["agents"]
actor_params = agents_config["actor"]
actor = AGENTS.get_from_params(
**actor_params,
env_spec=env_spec,
)
critic_params = agents_config["critic"]
critic = AGENTS.get_from_params(
**critic_params,
env_spec=env_spec,
)
algorithm = cls(
**config_["algorithm"],
actor=actor,
critic=critic,
)
return algorithm
def prepare_for_trainer(
cls,
env_spec: EnvironmentSpec,
config: Dict
) -> "AlgorithmSpec":
config_ = config.copy()
agents_config = config_["agents"]
actor_params = agents_config["actor"]
actor = AGENTS.get_from_params(
**actor_params,
env_spec=env_spec,
)
critic_params = agents_config["critic"]
critic = AGENTS.get_from_params(
**critic_params,
env_spec=env_spec,
)
action_space = env_spec.action_space
assert isinstance(action_space, Box)
action_boundaries = [
action_space.low[0],
action_space.high[0]
]
algorithm = cls(
**config_["algorithm"],
action_boundaries=action_boundaries,
actor=actor,
critic=critic,
)
return algorithm
def prepare_for_trainer(cls, config):
config_ = config.copy()
actor_state_shape = (
config_["shared"]["history_len"],
config_["shared"]["observation_size"],
)
actor_action_size = config_["shared"]["action_size"]
n_step = config_["shared"]["n_step"]
gamma = config_["shared"]["gamma"]
history_len = config_["shared"]["history_len"]
trainer_state_shape = (config_["shared"]["observation_size"], )
trainer_action_shape = (config_["shared"]["action_size"], )
actor_params = config_["actor"]
actor = AGENTS.get_from_params(**actor_params,
state_shape=actor_state_shape,
action_size=actor_action_size)
critic_params = config_["critic"]
critic = AGENTS.get_from_params(**critic_params,
state_shape=actor_state_shape,
action_size=actor_action_size)
n_critics = config_["algorithm"].pop("n_critics", 2)
critics = [
AGENTS.get_from_params(**critic_params,
state_shape=actor_state_shape,
action_size=actor_action_size)
for _ in range(n_critics - 1)
]
algorithm = cls(**config_["algorithm"],
actor=actor,
critic=critic,
critics=critics,
n_step=n_step,
gamma=gamma)
kwargs = {
"algorithm": algorithm,
"state_shape": trainer_state_shape,
"action_shape": trainer_action_shape,
"n_step": n_step,
"gamma": gamma,
"history_len": history_len
}
return kwargs
def prepare_for_sampler(cls, env_spec: EnvironmentSpec,
config: Dict) -> Union[ActorSpec, CriticSpec]:
config_ = config.copy()
agents_config = config_["agents"]
actor_params = agents_config["actor"]
actor = AGENTS.get_from_params(
**actor_params,
env_spec=env_spec,
)
return actor
def prepare_for_sampler(cls, config):
config_ = config.copy()
actor_state_shape = (
config_["shared"]["history_len"],
config_["shared"]["observation_size"],
)
actor_action_size = config_["shared"]["action_size"]
actor_params = config_["actor"]
actor = AGENTS.get_from_params(**actor_params,
state_shape=actor_state_shape,
action_size=actor_action_size)
history_len = config_["shared"]["history_len"]
kwargs = {"actor": actor, "history_len": history_len}
return kwargs