Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 318.00 +/- 163.17
+      value: 515.00 +/- 187.71
       name: mean_reward
       verified: false
 ---
@@ -62,17 +62,17 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 128),
-             ('buffer_size', 100500),
+OrderedDict([('batch_size', 256),
+             ('buffer_size', 100000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
              ('exploration_final_eps', 0.01),
              ('exploration_fraction', 0.1),
              ('frame_stack', 4),
              ('gradient_steps', 1),
-             ('learning_rate', 0.005),
+             ('learning_rate', 0.0001),
              ('learning_starts', 100000),
-             ('n_timesteps', 999000.0),
+             ('n_timesteps', 1000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),

args.yml

@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 3829660322
+    - 1494236837
   - - storage
     - null
   - - study_name

config.yml

@@ -1,8 +1,8 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 128
+    - 256
   - - buffer_size
-    - 100500
+    - 100000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -14,11 +14,11 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0.005
+    - 0.0001
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 999000.0
+    - 1000000.0
   - - optimize_memory_usage
     - false
   - - policy

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f73d1e2001805c3dc6f8c159f6cda8ca3ad157221ed062edc40af4b30695dfe2
-size 27218388
+oid sha256:6258cc27c5eb3ddc33f78ed87de6f245bfa72fa341c686c77e7761e5e9db3f78
+size 27218391

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92d850eee3cd8b71d7e85e62b80f6a8563e53842d1094f0fd87386848d5b0363
+oid sha256:3af1e2f7a3e80e3fc8564a384e83cf44625fcc697598e94ef4c5da12de3661f9
 size 13505739

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1161f246439e4189820f16991581b18b093e8f7a5c10df527a8b5dae4fc45e1c
+oid sha256:85ccbe0193b575a6efe5d5258f0b4129f7e17c8ff027f6a25050340348d0792b
 size 13504937

results.json

@@ -1 +1 @@
-{"mean_reward": 318.0, "std_reward": 163.17475294910057, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-05-15T15:16:42.379486"}
+{"mean_reward": 515.0, "std_reward": 187.70988253152788, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-05-15T18:19:28.614711"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43c29861b22b9c14b563a5da94071cbfb62b6a7fdca89f221e75b1985b09eb8b
-size 40656
+oid sha256:9b1c3bba59625df2dd82132d862925bdee9fa8120e7d1acb39b8ce85f30dee7b
+size 38381