{"env": {"name": "tensoraerospace.envs.b747_vec_torch.ImprovedB747VecEnvTorch", "params": {"num_envs": 64, "dt": 0.1, "tn": 20.0, "initial_state": [0.0, 0.0, 0.0, 0.0], "device": "cuda", "seed": 1, "auto_reset": true, "step_randomization": {"amplitude_deg_range": [1.0, 1.0], "min_abs_amplitude_deg": 1.0, "step_time_sec_range": [5.0, 5.0], "ref_change_threshold_rad": 0.0017453292519943296, "min_step_amp_rad": 0.008726646259971648}, "action_space": "Box(-1.0, 1.0, (1,), float32)", "observation_space": "Box(-1.0, 1.0, (4,), float32)"}}, "policy": {"name": "tensoraerospace.agent.ppo.model.PPO", "params": {"gamma": 0.995, "max_episodes": 90000, "rollout_len": 256, "clip_pram": 0.15, "num_epochs": 2, "batch_size": 16384, "entropy_coef": 0.01, "actor_lr": 0.0001, "critic_lr": 0.0002, "gae_lambda": 0.95, "max_grad_norm": 0.5, "target_kl": 0.01, "normalize_obs": false, "normalize_reward": true, "actor_hidden_dim": 256, "critic_hidden_dim": 256, "actor_log_std_min": -5.0, "actor_log_std_max": -1.5, "eval_freq": 10, "seed": 336699, "log_dir": "runs/b747_vec64_step", "save_best_model": true, "best_model_dir": "b747_vec64_step_best", "save_best_async": true}}}