Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 319.00 +/- 125.73
+      value: 851.50 +/- 275.82
       name: mean_reward
       verified: false
 ---
@@ -62,8 +62,8 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 16),
-             ('buffer_size', 50000),
+OrderedDict([('batch_size', 32),
+             ('buffer_size', 100000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
              ('exploration_final_eps', 0.01),
@@ -72,9 +72,9 @@ OrderedDict([('batch_size', 16),
              ('gradient_steps', 1),
              ('learning_rate', 0.0001),
              ('learning_starts', 100000),
-             ('n_timesteps', 1000000.0),
+             ('n_timesteps', 10000000.0),
              ('optimize_memory_usage', False),
-             ('policy', 'MlpPolicy'),
+             ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),
              ('train_freq', 4),
              ('normalize', False)])

args.yml

@@ -18,7 +18,7 @@
   - - hyperparams
     - null
   - - log_folder
-    - logs7/
+    - logs5/
   - - log_interval
     - -1
   - - max_total_trials
@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 270929130
+    - 429927090
   - - storage
     - null
   - - study_name

config.yml

@@ -1,8 +1,8 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 16
+    - 32
   - - buffer_size
-    - 50000
+    - 100000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -18,11 +18,11 @@
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 1000000.0
+    - 10000000.0
   - - optimize_memory_usage
     - false
   - - policy
-    - MlpPolicy
+    - CnnPolicy
   - - target_update_interval
     - 1000
   - - train_freq

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:904521a0ed48b661412f6fe9965b5dc1ef4544a432fb62d9adf3c7ef6a952f51
-size 29192260
+oid sha256:3163d2b05b45150250ae6b9d67e4bcabb405ac5876d5f5e298512881f6d927df
+size 27215583

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44de05d5d2ae57d78fbee4ee2b0dc266bc850f6efe27842b03c97abc6da4b712
-size 14492399
+oid sha256:cbb7f209e445a93699fab0a850a09274e4551090af88655e9def9f0aedcbc229
+size 13502953

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2cfae794674f7b4cc055afa73e53ce8c528dd608a47f5019317cad5dc9de88e6
-size 14491521
+oid sha256:bc5ed37177ee7376f5e65591125b7a47170bc0bbbe25ca99405f56182aac4215
+size 13504745

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -1,9 +1,9 @@
-- OS: Linux-5.15.0-1039-oracle-x86_64-with-glibc2.35 # 45-Ubuntu SMP Thu Jul 13 19:37:23 UTC 2023
-- Python: 3.10.6
+- OS: Linux-5.15.0-1038-oracle-x86_64-with-glibc2.31 # 44~20.04.1-Ubuntu SMP Thu Jun 22 16:36:07 UTC 2023
+- Python: 3.9.17
 - Stable-Baselines3: 2.0.0
-- PyTorch: 2.0.1+cu117
+- PyTorch: 1.11.0+cu102
 - GPU Enabled: False
-- Numpy: 1.25.1
+- Numpy: 1.21.2
 - Cloudpickle: 2.2.1
 - Gymnasium: 0.28.1
 - OpenAI Gym: 0.26.2

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8844acf903ec50ff6390ef5cf264eb4433418c5bcd7a256635242a718636974d
-size 240647
+oid sha256:d38caf32749738c5aff1cd187080caa787c003bb5d151b2508441c7dd9c190c1
+size 205711

results.json

@@ -1 +1 @@
-{"mean_reward": 319.0, "std_reward": 125.73384588089239, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-07-27T09:05:03.377842"}
+{"mean_reward": 851.5, "std_reward": 275.81742149472717, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-07-28T03:28:54.201893"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:922f003a28eed0e06c39318a20a57e5be0c163aad32b2552a8c7ca7ed60c2088
-size 43152
+oid sha256:77cae0c0b68ea162037f18c93c4114918fdcdf8f7240f63b377df8d22f2bdbcc
+size 276708