{ "problem": 4, "timesteps": 40000, "seed": 20260911, "device": "cuda", "directional_probability": 1.0, "wall_seconds": 235.8, "model": "output\\rl\\q4_ppo.zip", "init": "output/rl/q4_ppo_bc_init.zip", "episodes": 40, "success_rate": 0.9, "mean_steps": 4.45, "mean_episode_reward": 12.621703708323922, "successes": 36 }