gomoku_rl.policy.common module

gomoku_rl.policy.common.get_optimizer(cfg: DictConfig, params: Iterable[Parameter]) → Optimizer

gomoku_rl.policy.common.make_critic(cfg: DictConfig, device: device | str | int | None)

gomoku_rl.policy.common.make_dataset_naive(tensordict: TensorDict, batch_size: int) → Generator[TensorDict, None, None]

gomoku_rl.policy.common.make_dqn_actor(cfg: DictConfig, action_spec: TensorSpec, device: device | str | int | None)

gomoku_rl.policy.common.make_egreedy_actor(actor: TensorDictModule, action_spec: TensorSpec, eps_init: float = 1.0, eps_end: float = 0.1, annealing_num_steps: int = 1000)

gomoku_rl.policy.common.make_ppo_ac(cfg: DictConfig, action_spec: TensorSpec, device: device | str | int | None)

gomoku_rl.policy.common.make_ppo_actor(cfg: DictConfig, action_spec: TensorSpec, device: device | str | int | None)