PPOActorTrajectoryRuntime
Minimal trajectory-runtime collaborator required by the PPO actor step processor.
Methods
abandon_episode
python
abandon_episode(step: int, *, session: ActiveTrajectorySession) -> Nonefinish_episode
python
finish_episode(trajectory_file: EpisodeTrajectoryDataSource | None, episode_data: dict[str, float], step: int, *, outcome: EpisodeOutcome, session: ActiveTrajectorySession, episode_length: int | None = None) -> Nonesnapshot_potential
python
snapshot_potential(utility_key: str, field_names: list[str]) -> dict[tuple[int, ...], float]