Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: EnduroNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 3.40 +/- 4.15
+      value: 155.30 +/- 17.60
       name: mean_reward
       verified: false
 ---
@@ -63,7 +63,7 @@ python -m rl_zoo3.push_to_hub --algo dqn --env EnduroNoFrameskip-v4 -f logs/ -or
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 100),
+OrderedDict([('batch_size', 128),
              ('buffer_size', 100000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
@@ -74,7 +74,7 @@ OrderedDict([('batch_size', 100),
              ('gradient_steps', 1),
              ('learning_rate', 0.00011447658254850165),
              ('learning_starts', 20000),
-             ('n_timesteps', 100000.0),
+             ('n_timesteps', 1000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),

dqn-EnduroNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8dd948a45fbc88dff9d1de0528b2970b062b87795b6069556d6eb29e269c14d6
-size 27242497
+oid sha256:e7f7099c6b27e499cdc9940e4019456f78ec9235a0b0e51399066c102cd828d9
+size 27245398

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c73c922f4d35f4a57a35399ced7c77d1143409b2b4ec751356c1682ea26af2f
-size 363896
+oid sha256:db1fe135baf7c3dfcfc983087716daee0fa1147998b17afec7a56609041a8162
+size 384000

results.json

@@ -1 +1 @@
-{"mean_reward": 3.4, "std_reward": 4.1521078984053394, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-12-08T11:43:47.026256"}
+{"mean_reward": 155.3, "std_reward": 17.595738120351758, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-12-08T14:07:05.057188"}