Views
No views yet
LunarLander-v3. The model card uses LunarLander-v2 because the Hugging Face Deep RL Course Unit 8 checker still looks for that environment tag.1{
2 "anneal_lr": true,
3 "batch_size": 16384,
4 "capture_video": false,
5 "clip_coef": 0.2,
6 "clip_vloss": true,
7 "course_env_id": "LunarLander-v2",
8 "cuda": true,
9 "ent_coef": 0.01,
10 "env_id": "LunarLander-v3",
11 "eval_deterministic": true,
12 "eval_episodes": 10,
13 "eval_video_dir": null,
14 "evaluate": false,
15 "exp_name": "ppo_lunarlander",
16 "gae": true,
17 "gae_lambda": 0.98,
18 "gamma": 0.999,
19 "headless": false,
20 "learning_rate": 0.0003,
21 "load_model_path": "checkpoints/ppo_lunarlander.pt",
22 "logs_dir": "runs/LunarLander-v3__ppo_lunarlander__3__1779423167",
23 "max_grad_norm": 0.5,
24 "minibatch_size": 64,
25 "norm_adv": true,
26 "num_envs": 16,
27 "num_minibatches": 256,
28 "num_steps": 1024,
29 "package_dir": "hub_package",
30 "repo_id": "ZuzEL/LunarLander-v2",
31 "save_model_path": "checkpoints/ppo_lunarlander.pt",
32 "seed": 3,
33 "target_kl": null,
34 "torch_deterministic": true,
35 "total_timesteps": 1000000,
36 "track": false,
37 "update_epochs": 4,
38 "upload_to_hub": true,
39 "vf_coef": 0.5,
40 "video_fps": 30,
41 "wandb_entity": null,
42 "wandb_project_name": "cleanRL"
43}