def _log_creation(self):
logger.info("creating a TD3 agent")
logger.info("the actor has %d parameters",
count_parameters(self._actor))
logger.info("the critic1 has %d parameters",
count_parameters(self._critic1))
logger.info("the critic2 has %d parameters",
count_parameters(self._critic2))
def _log_creation(self):
logger.info("Creating a SAC agent")
logger.info("The actor has %d parameters".format(
count_parameters(self._actor)))
logger.info('The critic1 has %d parameters',
count_parameters(self._critic1))
logger.info('The critic2 has %d parameters',
count_parameters(self._critic2))
def _log_creation(self):
if self._config.is_chef:
logger.info('Creating a SAC agent')
logger.info('The actor has %d parameters',
count_parameters(self._actor))
logger.info('The critic1 has %d parameters',
count_parameters(self._critic1))
logger.info('The critic2 has %d parameters',
count_parameters(self._critic2))
def _log_creation(self):
if self._config.is_chef:
logger.info("creating a sac agent")
logger.info("the actor has %d parameters",
count_parameters(self._actor))
logger.info("the critic1 has %d parameters",
count_parameters(self._critic1))
logger.info("the critic2 has %d parameters",
count_parameters(self._critic2))
def __init__(self, config, ob_space, ac_space,
actor, critic):
super().__init__(config, ob_space)
self._ac_space = ac_space
# build up networks
self._actor = actor(config, ob_space, ac_space, config.tanh_policy)
self._old_actor = actor(config, ob_space, ac_space, config.tanh_policy)
self._critic = critic(config, ob_space)
self._network_cuda(config.device)
self._actor_optim = optim.Adam(self._actor.parameters(), lr=config.lr_actor)
self._critic_optim = optim.Adam(self._critic.parameters(), lr=config.lr_critic)
sampler = RandomSampler()
self._buffer = ReplayBuffer(['ob', 'ac', 'done', 'rew', 'ret', 'adv', 'ac_before_activation'],
config.buffer_size,
sampler.sample_func)
if config.is_chef:
logger.info('Creating a PPO agent')
logger.info('The actor has %d parameters', count_parameters(self._actor))
logger.info('The critic has %d parameters', count_parameters(self._critic))
def _log_creation(self):
logger.info('creating a DDPG agent')
logger.info('the actor has %d parameters', count_parameters(self._actor))
logger.info('the critic has %d parameters', count_parameters(self._critic))
def _log_creation(self):
logger.info("Creating a DQN agent")
logger.info("The DQN has %d parameters".format(
count_parameters(self._dqn)))