Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 223.50 +/- 89.39
+      value: 268.50 +/- 78.17
       name: mean_reward
       verified: false
 ---
@@ -62,21 +62,21 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 64),
-             ('buffer_size', 100000),
+OrderedDict([('batch_size', 128),
+             ('buffer_size', 101000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
              ('exploration_final_eps', 0.01),
              ('exploration_fraction', 0.1),
-             ('frame_stack', 4),
+             ('frame_stack', 5),
              ('gradient_steps', 1),
-             ('learning_rate', 0.001),
+             ('learning_rate', 0.03),
              ('learning_starts', 100000),
-             ('n_timesteps', 800000.0),
+             ('n_timesteps', 900000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),
-             ('train_freq', 4),
+             ('train_freq', 5),
              ('normalize', False)])
 ```
 

args.yml

@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 122165829
+    - 2359424619
   - - storage
     - null
   - - study_name

config.yml

@@ -1,8 +1,8 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 64
+    - 128
   - - buffer_size
-    - 100000
+    - 101000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -10,15 +10,15 @@
   - - exploration_fraction
     - 0.1
   - - frame_stack
-    - 4
+    - 5
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0.001
+    - 0.03
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 800000.0
+    - 900000.0
   - - optimize_memory_usage
     - false
   - - policy
@@ -26,4 +26,4 @@
   - - target_update_interval
     - 1000
   - - train_freq
-    - 4
+    - 5

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47db1150cdd059039a22520123f465ec108ba6a4cbe7fad0e9d07d4e187c2ad3
-size 27218381
+oid sha256:eebb804468132ec5da5194fa3e91a2faf3229085379e55d92d25203ae64c3350
+size 27299107

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b19263a327fe6a9b664437c52cf77407c7a33078bea1cccf04d6ae8ba39a7621
-size 13505739
+oid sha256:4e4e8e1547e024fb27953de879cef7aefda49b04e6c81fb2ba3697adf55093c0
+size 13522123

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a43008a3c7cb53adef164e80bbd0424b8a2d480e874101e74cd3b8451ead551
-size 13504937
+oid sha256:66ab046edc1867018c60a39335720001e38df131b9119d67b216694a68e5770f
+size 13521321

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -1,9 +1,9 @@
 - OS: Linux-5.15.109+-x86_64-with-glibc2.35 # 1 SMP Fri Jun 9 10:57:30 UTC 2023
-- Python: 3.10.6
+- Python: 3.10.12
 - Stable-Baselines3: 2.0.0
 - PyTorch: 2.0.1+cu118
 - GPU Enabled: True
-- Numpy: 1.22.4
+- Numpy: 1.23.5
 - Cloudpickle: 2.2.1
 - Gymnasium: 0.28.1
 - OpenAI Gym: 0.26.2

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:45c22dea565ea383fff3bc509d502adf0a52a1b963f654e167c3ac3df4c52175
-size 278865
+oid sha256:38a799ac9657faeb813344ec56ae265d381f35a430091e2c90337e271719a9db
+size 278674

results.json

@@ -1 +1 @@
-{"mean_reward": 223.5, "std_reward": 89.38819832617726, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-07-20T18:25:43.430477"}
+{"mean_reward": 268.5, "std_reward": 78.16808811785025, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-08-10T14:35:33.969344"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba3494e7d91cb5b40bef174b8d76a6fd17e2e50ccb7e4d67b256cdb3e2b3d346
-size 33730
+oid sha256:917e410675e577ba80b0fba6d9606ef82821fa1fb9e016da330a927526492397
+size 37830