step 8_520_000 . Checkpoint from initial model taken and trained further at a lower learning rate 2nd

README.md

@@ -16,7 +16,7 @@ model-index:
       type: Pixelcopter-PLE-v0
     metrics:
     - type: mean_reward
-      value: 258.10 +/- 236.89
+      value: 317.20 +/- 181.22
       name: mean_reward
       verified: false
 ---

ppo-Pixelcopter-PLE-v0.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c20f18b593a534772420a2aa30cffa89e5905a3e6ae0444e35ad3b2ffc00710
+oid sha256:28dd68ede13a1fbd598aa00e99e72ae8cde78f7f5b599987128005459acc6c67
 size 147853

results.json

@@ -1 +1 @@
-{"mean_reward": 258.1, "std_reward": 236.890459917659, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2023-01-22T14:31:52.925925"}
+{"mean_reward": 317.2, "std_reward": 181.21633480456447, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2023-01-22T14:37:29.680369"}