Method learn

stable_baselines3/sac/sac.py:304–320

(
        self: SelfSAC,
        total_timesteps: int,
        callback: MaybeCallback = None,
        log_interval: int = 4,
        tb_log_name: str = "SAC",
        reset_num_timesteps: bool = True,
        progress_bar: bool = False,
    )

            self.logger.record("train/ent_coef_loss", np.mean(ent_coef_losses))

    def learn(
        self: SelfSAC,
        total_timesteps: int,
        callback: MaybeCallback = None,
        log_interval: int = 4,
        tb_log_name: str = "SAC",
        reset_num_timesteps: bool = True,
        progress_bar: bool = False,
    ) -> SelfSAC:
        return super().learn(
            total_timesteps=total_timesteps,
            callback=callback,
            log_interval=log_interval,
            tb_log_name=tb_log_name,
            reset_num_timesteps=reset_num_timesteps,
            progress_bar=progress_bar,
        )

    def _excluded_save_params(self) -> list[str]:
        return super()._excluded_save_params() + ["actor", "critic", "critic_target"]  # noqa: RUF005