Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 243.00 +/- 68.89
+      value: 329.00 +/- 157.97
       name: mean_reward
       verified: false
 ---
@@ -62,17 +62,17 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 32),
-             ('buffer_size', 100000),
+OrderedDict([('batch_size', 16),
+             ('buffer_size', 150000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
              ('exploration_final_eps', 0.01),
              ('exploration_fraction', 0.1),
              ('frame_stack', 4),
              ('gradient_steps', 1),
-             ('learning_rate', 0.0001),
+             ('learning_rate', 2e-05),
              ('learning_starts', 100000),
-             ('n_timesteps', 1000000.0),
+             ('n_timesteps', 10000.0),
              ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),

args.yml

@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 279034446
+    - 942847196
   - - storage
     - null
   - - study_name

config.yml

@@ -1,8 +1,8 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 32
+    - 16
   - - buffer_size
-    - 100000
+    - 150000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -14,11 +14,11 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0.0001
+    - 2.0e-05
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 1000000.0
+    - 10000.0
   - - optimize_memory_usage
     - false
   - - policy

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:481b33fb2287f8279781fc27c278604edbe2e797676653378472ec8aa7ad9fa0
-size 27220209
+oid sha256:545447b52726e403bcce1b0f5a38c35718425185f1641aea1b005c33ef10c59c
+size 13711811

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43605d9f647f3b2f8c3f77a745efab22b04df8aee6d614cc94612c9b875823b0
-size 13506236
+oid sha256:0cc351647a59bfe65acdd23d06b233cc39b75fad5d3d7450f37fb5322659f8d2
+size 1120

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65a24ff5e4be7944e3a47800abe8cb153402ce9a6ade9f2b55af22e3e498b536
+oid sha256:caf52418779159e9a18f9eabfe9cacbdb8c4909e7c30119e5c5e265578ec94cc
 size 13505370

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cfaabae35da707cb869348b84f1441ad6abf2b64bea8609c6f76ae42b101f823
-size 268528
+oid sha256:d1220fd2663a829192a2b709c767eb2debf32246e6f8d3b5838a5840975ff8d0
+size 242623

results.json

@@ -1 +1 @@
-{"mean_reward": 243.0, "std_reward": 68.89121859859934, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-03-11T21:48:20.009994"}
+{"mean_reward": 329.0, "std_reward": 157.96835126062436, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-03-11T22:04:31.141072"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb6a8eded9c8b852caa90fabc6d2b8c7e603f5a72f95a3d94e5c7fd5285c38c1
-size 7516
+oid sha256:c922f99b683ff977decf29898d5df84f4c9d440b6eff971d94fbbcbc559cf0a4
+size 646