Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 128.00 +/- 49.86
+      value: 139.00 +/- 61.02
       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', 32),
-             ('buffer_size', 100000),
+OrderedDict([('batch_size', 128),
+             ('buffer_size', 1000000),
              ('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', 0.0001),
+             ('learning_rate', 1e-06),
              ('learning_starts', 100000),
-             ('n_timesteps', 1000000),
+             ('n_timesteps', 10000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),

args.yml

@@ -18,7 +18,7 @@
   - - hyperparams
     - null
   - - log_folder
-    - logs2/
+    - logs5/
   - - log_interval
     - -1
   - - max_total_trials
@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 1611369050
+    - 1497195639
   - - storage
     - null
   - - study_name

config.yml

@@ -1,8 +1,8 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 32
+    - 128
   - - buffer_size
-    - 100000
+    - 1000000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -14,11 +14,11 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0.0001
+    - 1.0e-06
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 1000000
+    - 10000000.0
   - - optimize_memory_usage
     - false
   - - policy

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c5192ed63a434797535d48ac9ccb25f9a7c15feeff655ca96813d00b987c6d96
-size 13709288
+oid sha256:31b7308934101deabc280e51f46d2a5ac5cd897c4d0cd171ed85b030d4f01701
+size 13709215

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b69a6cb94f8a9599ea4291f66c0618fbf541f7509cd0bdc82e64322acea359b
+oid sha256:81abf2e1628e006879e47273d60bdc52aeebf32546d72a76ed96ef912f50d023
 size 687

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0164dfb8103edfdcac15663573bfc80dc6585dcfc93ee4533252692be342175b
+oid sha256:9c18975707e17ad8562413b0601e0f31d192603aab1b37ecaf4e2c13eff2fcf1
 size 13504745

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -1,4 +1,4 @@
-- OS: Linux-5.15.0-1038-oracle-x86_64-with-glibc2.35 # 44-Ubuntu SMP Thu Jun 22 03:34:25 UTC 2023
+- OS: Linux-5.15.0-1039-oracle-x86_64-with-glibc2.35 # 45-Ubuntu SMP Thu Jul 13 19:37:23 UTC 2023
 - Python: 3.10.6
 - Stable-Baselines3: 2.0.0
 - PyTorch: 2.0.1+cu117

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a6976c90010e664a9463bc6bf7c177531305c6d63183da5236936acab10a589
-size 249457
+oid sha256:564c30ad614968e19b7da9d960f1767db1376e1c49e5c26df1fb357cfbed11da
+size 243037

results.json

@@ -1 +1 @@
-{"mean_reward": 128.0, "std_reward": 49.85980344927164, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-07-26T03:46:45.653388"}
+{"mean_reward": 139.0, "std_reward": 61.02458520956943, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-07-26T14:52:23.477170"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08f3f851d87058a026099e91007739f25163791f9d8b4c7b99e49bf7fe3ef36b
-size 268
+oid sha256:cda0e827fe9cdb5f59f1b71942d3af395c18612049de42510faccc7acb89059b
+size 225