def __init__(self, device: torch.device, settings: OnPolicyIterationReinforcerSettings, model: Model,
algo: AlgoBase, env_roller: EnvRollerBase) -> None:
self.device = device
self.settings = settings
self._trained_model = model.to(self.device)
self.env_roller = env_roller
self.algo = algo
def __init__(self, device: torch.device, settings: BufferedSingleOffPolicyIterationReinforcerSettings,
environment: gym.Env, model: Model, algo: AlgoBase, env_roller: ReplayEnvRollerBase):
self.device = device
self.settings = settings
self.environment = environment
self._trained_model = model.to(self.device)
self.algo = algo
self.env_roller = env_roller
def __init__(self, device: torch.device,
settings: BufferedMixedPolicyIterationReinforcerSettings,
env: VecEnv, model: Model, env_roller: ReplayEnvRollerBase,
algo: AlgoBase) -> None:
self.device = device
self.settings = settings
self.environment = env
self._trained_model = model.to(self.device)
self.env_roller = env_roller
self.algo = algo