{ "problem": 3, "timesteps": 40000, "seed": 20260911, "device": "cuda", "directional_probability": 0.0, "wall_seconds": 279.0, "model": "output\\rl\\q3_ppo.zip", "init": "output/rl/q3_ppo_bc_init.zip", "episodes": 40, "success_rate": 1.0, "mean_steps": 2.9, "mean_episode_reward": 17.57896344478182, "successes": 40 }