Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 6.50 +/- 10.74
+      value: 488.00 +/- 178.76
       name: mean_reward
       verified: false
 ---
@@ -71,9 +71,9 @@ OrderedDict([('batch_size', 32),
              ('exploration_fraction', 0.1),
              ('frame_stack', 4),
              ('gradient_steps', 1),
-             ('learning_rate', '0_1e-4'),
+             ('learning_rate', '1e-5_1e-4'),
              ('learning_starts', 100000),
-             ('n_timesteps', 1000),
+             ('n_timesteps', 1000000),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),

args.yml

@@ -56,7 +56,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 2627643323
+    - 3670464182
   - - storage
     - null
   - - study_name

config.yml

@@ -14,11 +14,11 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0_1e-4
+    - 1e-5_1e-4
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 1000
+    - 1000000
   - - optimize_memory_usage
     - false
   - - policy

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c142c1a54da05ae715249884bdea4223add19fefad3a3a5771ffe5082f553ef2
-size 13712338
+oid sha256:e41d9c4cd2fdc18c520483f84cf6ea80c9ef8335cb085e9ef9e5520bdcf38189
+size 27221459

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:215ea7d8898faa9284464c7109532cf36390d330fccab8d77eeba20628a32876
-size 1120
+oid sha256:36eba7ca50722c1108dbed94c9f0984cfc9dd10fd56c0ce78deedd5221d7244a
+size 13506172

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3a23165989abb38d3b56ef39567a66e5685db6d3bad7535da43f882b98b765f
+oid sha256:00728d54dafa54e67541b3f5950783276ec1b87185aedd727e243346039e571d
 size 13505370

results.json

@@ -1 +1 @@
-{"mean_reward": 6.5, "std_reward": 10.735455276791944, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2025-04-12T06:35:42.735938"}
+{"mean_reward": 488.0, "std_reward": 178.76241215647096, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2025-04-12T09:55:46.287874"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59b810eb78e86fa68251b7a1416483c0510098d1aa88022d85ac65f59310e2ab
-size 246
+oid sha256:707384018808c9d5fb63ec3799df8c2e3b1afde897756e8093495883fc009a8b
+size 40375