Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 15.50 +/- 12.54
+      value: 248.00 +/- 152.70
       name: mean_reward
       verified: false
 ---
@@ -72,7 +72,7 @@ OrderedDict([('batch_size', 32),
              ('gradient_steps', 1),
              ('learning_rate', 0.0001),
              ('learning_starts', 100000),
-             ('n_timesteps', 3000),
+             ('n_timesteps', 100000),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),

args.yml

@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 3727293767
+    - 1518036759
   - - storage
     - null
   - - study_name

config.yml

@@ -18,7 +18,7 @@
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 3000
+    - 100000
   - - optimize_memory_usage
     - false
   - - policy

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:51f5d6b06953c07e174547cc1391720b9de88dd41699acd6d53a6a97de1a0bd3
-size 13715773
+oid sha256:278de2d22d2b1ac390725a145845265598ea9a821d296d2378afb607fd57a4b5
+size 13719935

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d1d4a9bf585cd13022dfbcb34755fc97e4fc9d6c5a6e0c6c3cdf8968327f19a
-size 13504745
+oid sha256:23086112959614c2fbf8a047785670c22132308b2c5f778f79140e4e48d1bce3
+size 13504937

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -1,7 +1,7 @@
 - OS: Linux-5.15.90+-x86_64-with-debian-bullseye-sid # 1 SMP Thu Apr 6 11:02:12 UTC 2023
 - Python: 3.7.12
 - Stable-Baselines3: 1.8.0
-- PyTorch: 1.13.0+cpu
-- GPU Enabled: False
+- PyTorch: 1.13.0
+- GPU Enabled: True
 - Numpy: 1.21.6
 - Gym: 0.21.0

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a1336777fcc82df62976f19ade9aa131a3825f788d525d0e238ee57097cf4d0c
-size 272889
+oid sha256:79031709f4487b7f73de0aa07e830f658750d7f46e93203ceb8d1c9d528f4374
+size 267272

results.json

@@ -1 +1 @@
-{"mean_reward": 15.5, "std_reward": 12.539936203984453, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-04-13T02:44:55.545828"}
+{"mean_reward": 248.0, "std_reward": 152.6957759730111, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-04-13T02:53:34.283041"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c256140be5b6f442e2bb9b567c8a8491f27ceea5d2ecb1d36b20819e29ec5f4f
-size 288
+oid sha256:fcff594a913ff972b6f546ac5428c1918382051f437990424b4a1a96b61ed599
+size 5711