Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 599.00 +/- 121.79
+      value: 399.50 +/- 127.05
       name: mean_reward
       verified: false
 ---
@@ -62,7 +62,7 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 32),
+OrderedDict([('batch_size', 64),
              ('buffer_size', 100000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
@@ -70,12 +70,17 @@ OrderedDict([('batch_size', 32),
              ('exploration_fraction', 0.1),
              ('frame_stack', 4),
              ('gradient_steps', 1),
-             ('learning_rate', 0.0001),
+             ('learning_rate', 0.0002),
              ('learning_starts', 100000),
-             ('n_timesteps', 11000000.0),
+             ('n_timesteps', 1000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),
              ('train_freq', 4),
              ('normalize', False)])
 ```
+
+# Environment Arguments
+```python
+{'render_mode': 'rgb_array'}
+```

args.yml

@@ -2,7 +2,7 @@
 - - - algo
     - dqn
   - - conf_file
-    - null
+    - dqn.yml
   - - device
     - auto
   - - env
@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 3848301116
+    - 1335460079
   - - storage
     - null
   - - study_name

config.yml

@@ -1,6 +1,6 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 32
+    - 64
   - - buffer_size
     - 100000
   - - env_wrapper
@@ -14,11 +14,11 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0.0001
+    - 0.0002
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 11000000.0
+    - 1000000.0
   - - optimize_memory_usage
     - false
   - - policy

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:612196baca39bab6d418b7e9d1bafad4e93e993d4a74f05f1c5de99935932aca
-size 27224876
+oid sha256:a200d78f0272563f554302669389a2f407c2defe8b35477311f8660eccae593a
+size 27220132

dqn-SpaceInvadersNoFrameskip-v4/_stable_baselines3_version

@@ -1 +1 @@
-1.8.0
+2.2.1

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f423c7a37c53d8e6b8e6a5eb95124ce2ef66c18cb5a3dab89a833d6d0cafe95d
-size 13505739
+oid sha256:5a689a76a677a91b975bad6fe04ae3a49cd7ecce5a3a802a1d571f3bc7a2662a
+size 13506236

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f281ee85fc94c001f0e602de9ba538153a6c930428cd44c398138057cdd64467
-size 13504937
+oid sha256:b1fe8f3faa1ec60803de7692f75b4cd551e4a1941629cac1b6526cb080e0142d
+size 13505370

dqn-SpaceInvadersNoFrameskip-v4/pytorch_variables.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d030ad8db708280fcae77d87e973102039acd23a11bdecc3db8eb6c0ac940ee1
-size 431
+oid sha256:0c35cea3b2e60fb5e7e162d3592df775cd400e575a31c72f359fb9e654ab00c5
+size 864

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -1,7 +1,9 @@
-- OS: Linux-5.15.107+-x86_64-with-glibc2.31 # 1 SMP Sat Apr 29 09:15:28 UTC 2023
-- Python: 3.10.11
-- Stable-Baselines3: 1.8.0
-- PyTorch: 2.0.1+cu118
+- OS: Linux-6.1.58+-x86_64-with-glibc2.35 # 1 SMP PREEMPT_DYNAMIC Sat Nov 18 15:31:17 UTC 2023
+- Python: 3.10.12
+- Stable-Baselines3: 2.2.1
+- PyTorch: 2.1.0+cu121
 - GPU Enabled: True
-- Numpy: 1.22.4
-- Gym: 0.21.0
+- Numpy: 1.23.5
+- Cloudpickle: 2.2.1
+- Gymnasium: 0.29.1
+- OpenAI Gym: 0.26.2

env_kwargs.yml

@@ -1 +1 @@
-{}
+render_mode: rgb_array

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abbf8d01ffa14b876af0dc4bf8c97e90c7fe2bc24923b0bc7f9f3fe085d848c0
-size 203488
+oid sha256:42e6a8381812642bd620cd816395f30d5fe6088617a8b8dd4f4dee2d48998a8c
+size 255888

results.json

@@ -1 +1 @@
-{"mean_reward": 599.0, "std_reward": 121.79490958164057, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-05-26T20:46:38.323634"}
+{"mean_reward": 399.5, "std_reward": 127.05215464524795, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-01-04T10:59:58.828099"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5eb5a43329e82b41615e878edea174cb68563eb7cb7e54187ce4ecad33871beb
-size 43614
+oid sha256:b75bcc8465e890e90b8f0d6ecf9c0e97035310c1758764a2fa5d743bb63d86c6
+size 35097