PPO LunarLander-v2 3,000,000 training steps agent model card

README.md

@@ -16,7 +16,7 @@ model-index:
       type: LunarLander-v2
     metrics:
     - type: mean_reward
-      value: 272.45 +/- 35.84
+      value: 284.14 +/- 16.90
       name: mean_reward
       verified: false
 ---

ppo-LunarLander-v2-3e6-ppo-steps.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11d5a0a236386db2643657bd5cf8c283ae7841343219642e26e8f4c5e58c83e2
+oid sha256:5546604a58233a34a1c1daf46806592c88907501d4df8acfc9fbdff8b7692ddb
 size 146587

results.json

@@ -1 +1 @@
-{"mean_reward": 272.4462303134669, "std_reward": 35.83715995643372, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2022-12-31T16:35:53.956909"}
+{"mean_reward": 284.13705458332186, "std_reward": 16.898540983861096, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2022-12-31T16:38:00.847804"}