Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 542.00 +/- 152.14
+      value: 607.00 +/- 170.22
       name: mean_reward
       verified: false
 ---
@@ -57,20 +57,20 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 64),
-             ('buffer_size', 120000),
+OrderedDict([('batch_size', 128),
+             ('buffer_size', 130000),
              ('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', 7e-05),
+             ('learning_rate', 8e-05),
              ('learning_starts', 100000),
-             ('n_timesteps', 900000.0),
+             ('n_timesteps', 2000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
-             ('target_update_interval', 900),
+             ('target_update_interval', 980),
              ('train_freq', 4),
              ('normalize', False)])
 ```

args.yml

@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 2699922974
+    - 2190085864
   - - storage
     - null
   - - study_name

config.yml

@@ -1,8 +1,8 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 64
+    - 128
   - - buffer_size
-    - 120000
+    - 130000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -14,16 +14,16 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 7.0e-05
+    - 8.0e-05
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 900000.0
+    - 2000000.0
   - - optimize_memory_usage
     - false
   - - policy
     - CnnPolicy
   - - target_update_interval
-    - 900
+    - 980
   - - train_freq
     - 4

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b5dcee4abc1be008b849b801877de0231a0fd40c3579f59102f3340bfc2d3bb
-size 27224790
+oid sha256:3f4cecc0366bcc0be0fd795a3900894be5d7939d5906baa0595005a663f566d1
+size 27224794

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a1083c25fd1e80e1f96ae86cb63b80c6ed3e6dfb085404724c2c3484a4376b05
+oid sha256:3346f1dce4922633d8cd79da133b71843fa4abf251ac90904f050e992b54b1f8
 size 13505739

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9afb9fc4ab5f1ec30d02fe9dbc4e900bef9230476ed54fbcbd8d830998828ac
+oid sha256:bb075ce9f561a250c7c7d21617d26ec2c8e983c651bb5d374dae3784f79c17b5
 size 13504937

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:373cf67558da5b74e3054e99f006e9c50eabce3d834756ea670fd10532c989f9
-size 197209
+oid sha256:cef49b39f18b35b9738ebb38a966a841853746b32b159740f2ef103c8ae0464d
+size 195721

results.json

@@ -1 +1 @@
-{"mean_reward": 542.0, "std_reward": 152.1380951635717, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-29T15:36:12.285870"}
+{"mean_reward": 607.0, "std_reward": 170.2233826476257, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-30T13:01:28.879243"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7dd11f7dc9329365aecafdfbe5bd322f48dd202bf8dbd788d9a020cec3b19c3
-size 34949
+oid sha256:e64baf8f7d995312d22faf3a009535a097f2124802636eff3cda96b9d4b94396
+size 67396