Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 29.00 +/- 64.30
+      value: 711.00 +/- 224.01
       name: mean_reward
       verified: false
 ---
@@ -70,9 +70,9 @@ OrderedDict([('batch_size', 32),
              ('exploration_fraction', 0.1),
              ('frame_stack', 4),
              ('gradient_steps', 1),
-             ('learning_rate', 0.0005),
+             ('learning_rate', 0.0002),
              ('learning_starts', 100000),
-             ('n_timesteps', 100000.0),
+             ('n_timesteps', 1000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),

args.yml

@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 114831432
+    - 4580660
   - - storage
     - null
   - - study_name

config.yml

@@ -14,11 +14,11 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0.0005
+    - 0.0002
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 100000.0
+    - 1000000.0
   - - optimize_memory_usage
     - false
   - - policy

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b4a637547e2320949a74768539426f3bea68fadd0ee69b7e3a425fdfc48e25a
-size 13714997
+oid sha256:63f61fd87f083bb3f212979931cb384ab68677ea4005a461cb56d49a5cde8c11
+size 27220122

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5abe8db2e07b1bf0a893ef3b3eb2235d6e756cb21498d756ac4bc6d486ecacff
-size 1120
+oid sha256:77e0960d3c9420b6153932060ee1e55acc1ec80e02397820fc4d632b4683433e
+size 13506236

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1eb1a7caf01747d9f3561eef03e548ec52bb0a12918c0da397db6f7ee7e70cff
+oid sha256:e87e628c816cf40afa986c84a53f41fe1a9bf5f3b0fa30ad1712b898c4fc37bc
 size 13505370

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cae3de750f432875df42a4011aeedbc220ba38b313576b27b5dbc3ebf8a98d59
-size 294896
+oid sha256:75329840b7dce42a56fba4f40c60b99535d5a2a92b371d3b6114485f6ee2549c
+size 206158

results.json

@@ -1 +1 @@
-{"mean_reward": 29.0, "std_reward": 64.29618962271404, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-10-28T18:35:50.789835"}
+{"mean_reward": 711.0, "std_reward": 224.00669632848033, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-10-29T16:12:28.012089"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe79765c43033ff44853e414f534eaeb529e5a3b0884f24679a4f6e5c707fea9
-size 5795
+oid sha256:eb08933ce857d85f59ea0fa7a5be67da80f56eb208a4e9952df0fba13abb0896
+size 34415