Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 347.50 +/- 178.86
+      value: 401.50 +/- 239.43
       name: mean_reward
       verified: false
 ---
@@ -62,21 +62,21 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 64),
+OrderedDict([('batch_size', 1024),
              ('buffer_size', 100000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
              ('exploration_final_eps', 0.01),
-             ('exploration_fraction', 0.1),
+             ('exploration_fraction', 0.025),
              ('frame_stack', 4),
-             ('gradient_steps', 496),
-             ('learning_rate', 0.0003),
-             ('learning_starts', 0),
-             ('n_timesteps', 1015808),
+             ('gradient_steps', 1),
+             ('learning_rate', 0.0001),
+             ('learning_starts', 10000),
+             ('n_timesteps', 1000000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
-             ('target_update_interval', -1),
-             ('train_freq', 1024),
+             ('target_update_interval', 1000),
+             ('train_freq', 4),
              ('normalize', False)])
 ```
 

args.yml

@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 160830502
+    - 265090744
   - - storage
     - null
   - - study_name

config.yml

@@ -1,6 +1,6 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 128
+    - 1024
   - - buffer_size
     - 100000
   - - env_wrapper
@@ -16,7 +16,7 @@
   - - learning_rate
     - 0.0001
   - - learning_starts
-    - 1000000.0
+    - 10000
   - - n_timesteps
     - 1000000.0
   - - optimize_memory_usage

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c1f6282d1c0d2906cdf6d57f71f280c5279354fbdf817a35680adfdf86d0520
-size 13715027
+oid sha256:631f377061bbe104afe5783aeef3d9f4baec192b586baecf92c644feb1c4b1da
+size 27220145

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3fc84919e676e2bdcaa04814c7dfa2b7625d8d53a818418c6db25771364db93
-size 1120
+oid sha256:6087e71397fd9c75ee574d9f4cc334904c0e051a0cc44f33b33eeb2c7946070b
+size 13506236

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c38d865f86856ed2fe47e0e02126da9d99c8a33e6b26026f1e2c51fe4c5bc42
+oid sha256:800b917b9a9178db8a3d33c4fa92e74b2a86a375b8b5e6dce9586943d8b968f9
 size 13505370

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c58c056863b211fa6ef5376d192cd13a5816b5caabe7059aa53016f986040479
-size 274469
+oid sha256:767c4e20c0d9b64b4c168ca2032209c82a40398f96ac64aeb21439829410c005
+size 245193

results.json

@@ -1 +1 @@
-{"mean_reward": 257.0, "std_reward": 38.80721582386451, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-02-27T10:00:58.942785"}
+{"mean_reward": 401.5, "std_reward": 239.4269199568002, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-02-27T15:17:29.785296"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48cec9db3c8bd7e57f87828dc6511a609f586f72d88a9ec540d263cc477fb27a
-size 48435
+oid sha256:7119cd1f7a617469a4b69042d57e7b993518b3c08264157df6f162f60144b5b4
+size 41098