PPO LunarLander-v2 trained agent with 1 million steps

README.md

@@ -16,7 +16,7 @@ model-index:
       type: LunarLander-v2
     metrics:
     - type: mean_reward
-      value: 223.15 +/- 24.68
+      value: 232.80 +/- 16.49
       name: mean_reward
       verified: false
 ---

ppo-LunarLander-v2.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9581a43913da535f3753f323cee2f88f0defae3cb7866575cc67e19751888663
+oid sha256:e7b4e312b3992b5f1735fad3c83938ca39fd8b390ed1818bc984ec9df852cc39
 size 148678

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2df9b9529f1a3d8ec907a9c96c11ca9ca69fcf361e4e8df594a0dc8370e29b35
-size 157992
+oid sha256:63bf40c86cda734c7dc81bec20c2dab8ddfba09b80c8f49e083b6b779bc0a246
+size 163343

results.json

@@ -1 +1 @@
-{"mean_reward": 223.15379074659396, "std_reward": 24.678098244155432, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2025-05-21T19:50:13.776055"}
+{"mean_reward": 232.800053, "std_reward": 16.492303245730188, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2025-05-21T19:53:42.397972"}