Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 571.50 +/- 182.83
+      value: 690.00 +/- 137.15
       name: mean_reward
       verified: false
 ---
@@ -58,16 +58,16 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 ## Hyperparameters
 ```python
 OrderedDict([('batch_size', 32),
-             ('buffer_size', 100000),
+             ('buffer_size', 120000),
              ('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', 0.00015),
              ('learning_starts', 100000),
-             ('n_timesteps', 1000000.0),
+             ('n_timesteps', 3000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),

args.yml

@@ -18,7 +18,7 @@
   - - hyperparams
     - null
   - - log_folder
-    - logs/
+    - content/drive/MyDrive/logs
   - - log_interval
     - -1
   - - max_total_trials
@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 896005373
+    - 295434025
   - - storage
     - null
   - - study_name

config.yml

@@ -2,7 +2,7 @@
 - - - batch_size
     - 32
   - - buffer_size
-    - 100000
+    - 120000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -14,11 +14,11 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0.0001
+    - 0.00015
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 1000000.0
+    - 3000000.0
   - - optimize_memory_usage
     - false
   - - policy

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37bfe913b0a250a807117a693bb9178fd033a94b27065f52673b75ce196c7862
-size 27224788
+oid sha256:8ae9ad0217a4a60e3d4be1b77dde9f0834d5d92607fe1ed2c09ae7a1879e9eba
+size 27224794

dqn-SpaceInvadersNoFrameskip-v4/_stable_baselines3_version

@@ -1 +1 @@
-1.6.2
+1.7.0a10

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4390e4ebf1b0f5fb6e228ae073f904d295423f1bce5779cda878bccbc5e98d39
+oid sha256:1502a4590c32e1c137ae4f7e16e8f9dfb2eae1f0a21a5e08497c0ce767f0a1c7
 size 13505739

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7a038fe18dc0d284888c5882883b5f9de6fd859963cb248d5dcc7604d919b6a
+oid sha256:93a8854450947cd8c4405a35f67770ea5c22e4d52874d121693795ab05af50e5
 size 13504937

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -1,6 +1,6 @@
 OS: Linux-5.10.133+-x86_64-with-glibc2.27 #1 SMP Fri Aug 26 08:44:51 UTC 2022
 Python: 3.8.16
-Stable-Baselines3: 1.6.2
+Stable-Baselines3: 1.7.0a10
 PyTorch: 1.13.0+cu116
 GPU Enabled: True
 Numpy: 1.21.6

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86213e6fbab24982dfe0a1602e29dcc8faffb6935df9c0e5c59f3014ad0e7963
-size 244536
+oid sha256:593071b8860d99b8a6de71327cac6033f91ab512a745f742a6a3bed87849c0cb
+size 220610

results.json

@@ -1 +1 @@
-{"mean_reward": 571.5, "std_reward": 182.8257367002797, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-20T14:13:41.785070"}
+{"mean_reward": 690.0, "std_reward": 137.14955340794953, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-23T17:47:30.715187"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f7045365a1bcd2357389ae8ea3433ef3c4fb72b780f8f3a3f12a68512c9bbf1
-size 37019
+oid sha256:5c665bf32867ac5d7026262c3c96566ad0104ae0b8725460a28f33fde3fabe53
+size 90824