config.json
1.3 KB · 44 lines · json Raw
1 {
2 "algorithm": "PPO",
3 "library": "stable-baselines3",
4 "base_environment": "LunarLander-v3",
5 "model_filename": "ppo-LunarLander-v3-flip-128x128.zip",
6 "reward_version": "v5-soft-touchdown-refinement",
7 "reward_config_filename": "reward_config.json",
8 "reward_wrapper_filename": "flip_landing_reward_wrapper.py",
9 "observation_dimensions": 11,
10 "architecture": {
11 "actor_layers": [
12 128,
13 128
14 ],
15 "critic_layers": [
16 128,
17 128
18 ]
19 },
20 "evaluation": {
21 "n_episodes": 100,
22 "mean_shaped_reward": -413.80892156538437,
23 "std_shaped_reward": 213.20887834470014,
24 "shaped_course_score": -627.0177999100845,
25 "min_shaped_reward": -1183.712121425813,
26 "max_shaped_reward": -294.95529900279223,
27 "mean_original_reward": -671.0025925747099,
28 "std_original_reward": 62.16580799530145,
29 "original_course_score": -733.1684005700113,
30 "min_original_reward": -751.9068113021244,
31 "max_original_reward": -397.9538147700128,
32 "original_reward_200_rate": 0.0,
33 "flip_completion_rate": 0.15,
34 "recovery_completion_rate": 0.09,
35 "recovery_given_flip_rate": 0.6,
36 "safe_landing_rate": 0.0,
37 "flip_landing_rate": 0.0,
38 "landing_given_flip_rate": 0.0,
39 "stable_landing_rate": 0.0,
40 "inside_landing_zone_rate": 0.44,
41 "outside_zone_landing_rate": 0.0,
42 "in_zone_crash_rate": 0.13
43 }
44 }