def _initialize(self):
"""Initialize non-common things."""
if not self.args.test:
# load demo replay memory
demos = self._load_demos()
if self.use_n_step:
demos, demos_n_step = common_utils.get_n_step_info_from_demo(
demos, self.hyper_params["N_STEP"],
self.hyper_params["GAMMA"])
self.memory_n = ReplayBuffer(
buffer_size=self.hyper_params["BUFFER_SIZE"],
n_step=self.hyper_params["N_STEP"],
gamma=self.hyper_params["GAMMA"],
demo=demos_n_step,
)
# replay memory
self.beta = self.hyper_params["PER_BETA"]
self.memory = PrioritizedReplayBuffer(
self.hyper_params["BUFFER_SIZE"],
self.hyper_params["BATCH_SIZE"],
demo=demos,
alpha=self.hyper_params["PER_ALPHA"],
epsilon_d=self.hyper_params["PER_EPS_DEMO"],
)
def _initialize(self):
"""Initialize non-common things."""
self.use_n_step = self.hyper_params["N_STEP"] > 1
if not self.args.test:
# load demo replay memory
with open(self.args.demo_path, "rb") as f:
demos = pickle.load(f)
if self.use_n_step:
demos, demos_n_step = common_utils.get_n_step_info_from_demo(
demos, self.hyper_params["N_STEP"], self.hyper_params["GAMMA"]
)
# replay memory for multi-steps
self.memory_n = NStepTransitionBuffer(
buffer_size=self.hyper_params["BUFFER_SIZE"],
n_step=self.hyper_params["N_STEP"],
gamma=self.hyper_params["GAMMA"],
demo=demos_n_step,
)
# replay memory
self.beta = self.hyper_params["PER_BETA"]
self.memory = PrioritizedReplayBufferfD(
self.hyper_params["BUFFER_SIZE"],
self.hyper_params["BATCH_SIZE"],
demo=demos,
alpha=self.hyper_params["PER_ALPHA"],
epsilon_d=self.hyper_params["PER_EPS_DEMO"],
)
