Upload PPO LunarLander-v2 trained agent, 1st try, 1000000 steps

README.md

@@ -10,7 +10,7 @@ model-index:
   results:
   - metrics:
     - type: mean_reward
-      value: 286.30 +/- 20.55
+      value: 271.34 +/- 24.37
       name: mean_reward
     task:
       type: reinforcement-learning

ppo-LunarLander-v2b.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bac1579a22be0dffa2b6667cb66d07f0b5825da14caf740b3c46b1d72143d5ef
+oid sha256:65e5dd72846d1dd8bfa239a470a54ed5dd81c884304d12870d59c3b37d8477f7
 size 147637

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8f7cf83a08d00c130db6b3c3ad024334060d7cea69da2aff02da6429eaee3a1
-size 183859
+oid sha256:db18d7a69a502dcb28cfab038abfb74bbbffc68be49599dc0ac088cbbe55fefe
+size 185300

results.json

@@ -1 +1 @@
-{"mean_reward": 286.30428317564173, "std_reward": 20.547091244442004, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2022-05-08T04:19:23.170870"}
+{"mean_reward": 271.3413132432498, "std_reward": 24.36661590399868, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2022-05-08T04:20:05.843005"}