src.model.reinforcement.ActorCritic module

class src.model.reinforcement.ActorCritic.ACModelHandler(*args, **kwargs)

Bases: RLModelHandler

ReinforceModelHandler.

Generate and manage a reinforcement model

calculate_reward(flags: Tensor, expected_classes: Tensor, expected_qoe: Tensor) → Tuple[int, Tuple[int, ...], Dict[str, Any] | None]

execute_episode(dump_memory: bool = True, train: bool = True) → None

get_action(current_state: Tensor) → Tuple[int, List[float], float]

train_epoch(*args, **kwargs)