Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 244.00 +/- 135.66
+      value: 373.00 +/- 91.14
       name: mean_reward
       verified: false
 ---
@@ -63,22 +63,22 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 64),
-             ('buffer_size', 200000),
+OrderedDict([('batch_size', 16),
+             ('buffer_size', 100000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
-             ('exploration_final_eps', 0.0944),
-             ('exploration_fraction', 0.137),
-             ('exploration_initial_eps', 0.749),
+             ('exploration_final_eps', 0.0316),
+             ('exploration_fraction', 0.159),
+             ('exploration_initial_eps', 0.85),
              ('frame_stack', 4),
-             ('gamma', 0.969),
+             ('gamma', 0.914),
              ('gradient_steps', 1),
-             ('learning_rate', 4.53e-06),
-             ('learning_starts', 100),
+             ('learning_rate', 1.95e-06),
+             ('learning_starts', 1000),
              ('n_timesteps', 1000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
-             ('target_update_interval', 11000),
+             ('target_update_interval', 8000),
              ('train_freq', 4),
              ('normalize', False)])
 ```

args.yml

@@ -56,7 +56,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 3722634297
+    - 2600369761
   - - storage
     - null
   - - study_name

config.yml

@@ -1,26 +1,26 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 64
+    - 16
   - - buffer_size
-    - 200000
+    - 100000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
-    - 0.0944
+    - 0.0316
   - - exploration_fraction
-    - 0.137
+    - 0.159
   - - exploration_initial_eps
-    - 0.749
+    - 0.85
   - - frame_stack
     - 4
   - - gamma
-    - 0.969
+    - 0.914
   - - gradient_steps
     - 1
   - - learning_rate
-    - 4.53e-06
+    - 1.95e-06
   - - learning_starts
-    - 100
+    - 1000
   - - n_timesteps
     - 1000000.0
   - - optimize_memory_usage
@@ -28,6 +28,6 @@
   - - policy
     - CnnPolicy
   - - target_update_interval
-    - 11000
+    - 8000
   - - train_freq
     - 4

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4bd97e1b878818ab80543560d39b07ee5e12cf134a9001ef96f63b8e0353527
-size 27219628
+oid sha256:c61c3667160b30cd1a78c93f42f2615822aa8ec543137712de86841eb4d2f8c7
+size 27219626

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33c16f81428a1c8f1ccdd6bb89b23810dc6060372296deb2dd4c5c04fedf9dac
+oid sha256:3fc4f69dd17396b0384a06b50e71859359fae8ade75e8ed400cea4031cfdecf0
 size 13506569

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f7837ddb07fe00003e55b5b52f4bc465bf2338ff1eda2f90d60b66d85ab1166
+oid sha256:c712be936d1364d714b27be4383f89857077d2ce7143ccc318273dd03f377cb6
 size 13505767

results.json

@@ -1 +1 @@
-{"mean_reward": 244.0, "std_reward": 135.66134305689297, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2026-01-18T15:13:47.914739"}
+{"mean_reward": 373.0, "std_reward": 91.13725912051558, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2026-01-18T18:51:29.067018"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbeecb2f9351bc7a9654efcc6a2891ebca0804107f69e2adfa4a1236f98dd6a9
-size 44777
+oid sha256:869622cb3086aba58d119f629ea2502d855115e83c522e61a8105a038621f3f3
+size 40800