Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 274.50 +/- 31.50
+      value: 720.50 +/- 297.87
       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', 8),
-             ('buffer_size', 100000),
+OrderedDict([('batch_size', 64),
+             ('buffer_size', 120000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
              ('exploration_final_eps', 0.01),
-             ('exploration_fraction', 0.2),
+             ('exploration_fraction', 0.05),
              ('frame_stack', 4),
              ('gradient_steps', 1),
-             ('learning_rate', 0.01),
+             ('learning_rate', 0.0001),
              ('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
-    - 1459184332
+    - 614042303
   - - storage
     - null
   - - study_name

config.yml

@@ -1,24 +1,24 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 8
+    - 64
   - - buffer_size
-    - 100000
+    - 120000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
     - 0.01
   - - exploration_fraction
-    - 0.2
+    - 0.05
   - - frame_stack
     - 4
   - - gradient_steps
     - 1
   - - learning_rate
-    - 0.01
+    - 0.0001
   - - 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 sha256:0f00a9d537e9deb127eba8e413e9b2a69ab581f10c3d8a4ae16e2aa5f9f94263
-size 13719741
+oid sha256:abf543cdd48f4f84d3f60578e2f6815e4c379ed3f0d44f726e0e1767254ad3b6
+size 27224804

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f76c5127824678f0941ffa60d518bcb749b16de75e8f608b27412859db8df6c
-size 687
+oid sha256:bb89fc61d66db36ab6ffaa515608223fde97ea744685785f39e46d1a34a51ab7
+size 13505739

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f73eba39c459fd249b3a798826b22272ce5d35a6a9ee6238dcaf1f3fe2768f3
+oid sha256:772eb9e5d5882ad8cd73bf4365d368dbfabd926bda940e020809a4bd80f2254e
 size 13504937

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -1,5 +1,5 @@
-- OS: Linux-5.10.147+-x86_64-with-glibc2.27 # 1 SMP Sat Dec 10 16:00:40 UTC 2022
-- Python: 3.8.16
+- OS: Linux-5.10.147+-x86_64-with-glibc2.29 # 1 SMP Sat Dec 10 16:00:40 UTC 2022
+- Python: 3.8.10
 - Stable-Baselines3: 1.7.0
 - PyTorch: 1.13.1+cu116
 - GPU Enabled: True

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e13a11a839fea521ecfd34a27ae00521e04cfb8dd63fb3e005e2d701b498e2b
-size 261744
+oid sha256:80fb91beefb9e169d9d2ea5423dd939fea55c2ad85109720eb68140f427d393e
+size 178350

results.json

@@ -1 +1 @@
-{"mean_reward": 274.5, "std_reward": 31.5, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-01-12T14:11:42.318452"}
+{"mean_reward": 720.5, "std_reward": 297.8711969962856, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-01-23T22:37:27.330071"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98ca07ac91e458e903826c179ce59cf45fb2f6b9bf20c3014e16a8b873d7dd67
-size 5903
+oid sha256:687e61cadb4fa973e1eccf54e983b2f1a42844316d1ebfed1d4ddead7d5c9c90
+size 36151