Trained for 500 000 steps with 16 env

README.md

@@ -10,7 +10,7 @@ model-index:
   results:
   - metrics:
     - type: mean_reward
-      value: 0.90 +/- 0.30
+      value: 0.80 +/- 0.40
       name: mean_reward
     task:
       type: reinforcement-learning

ppo-FrozenLake-v1.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96039abba8eb14bfe52cb40d205055102f4177141ebc558fa4298b7464b5e42d
+oid sha256:de3b29db70f892c794691b7a9e701501aaf639302c1a6c7305240554cae2c65e
 size 153712

results.json

@@ -1 +1 @@
-{"mean_reward": 0.9, "std_reward": 0.30000000000000004, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2022-06-07T12:45:28.942375"}
+{"mean_reward": 0.8, "std_reward": 0.4, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2022-06-07T12:47:59.000625"}