Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 162.50 +/- 94.74
+      value: 319.00 +/- 125.73
       name: mean_reward
       verified: false
 ---
@@ -62,19 +62,19 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 128),
-             ('buffer_size', 1000000),
+OrderedDict([('batch_size', 16),
+             ('buffer_size', 50000),
              ('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', 1e-06),
+             ('learning_rate', 0.0001),
              ('learning_starts', 100000),
-             ('n_timesteps', 10000000.0),
+             ('n_timesteps', 1000000.0),
              ('optimize_memory_usage', False),
-             ('policy', 'CnnPolicy'),
+             ('policy', 'MlpPolicy'),
              ('target_update_interval', 1000),
              ('train_freq', 4),
              ('normalize', False)])

args.yml

@@ -18,7 +18,7 @@
   - - hyperparams
     - null
   - - log_folder
-    - logs6/
+    - logs7/
   - - log_interval
     - -1
   - - max_total_trials
@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 3305755252
+    - 270929130
   - - storage
     - null
   - - study_name

config.yml

@@ -1,8 +1,8 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 128
+    - 16
   - - buffer_size
-    - 1000000
+    - 50000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -14,15 +14,15 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 1.0e-06
+    - 0.0001
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 10000000.0
+    - 1000000.0
   - - optimize_memory_usage
     - false
   - - policy
-    - CnnPolicy
+    - MlpPolicy
   - - target_update_interval
     - 1000
   - - train_freq

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0fc8c87806d4c68d49289814aa79a636065d27f9c78728f5abd2896e9bcc176
-size 27218054
+oid sha256:904521a0ed48b661412f6fe9965b5dc1ef4544a432fb62d9adf3c7ef6a952f51
+size 29192260

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1384fb70056808fd49020e21dd1c48eaa4ebdc1223d9bcf300d22ab998385d9e
-size 13505419
+oid sha256:44de05d5d2ae57d78fbee4ee2b0dc266bc850f6efe27842b03c97abc6da4b712
+size 14492399

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e51c0e845ccdaf52e921edbef01d7244ae8a0773777fd10b5ac4c490c0be8d5
-size 13504745
+oid sha256:2cfae794674f7b4cc055afa73e53ce8c528dd608a47f5019317cad5dc9de88e6
+size 14491521

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:459eb5214cce8826056ab0d83fbf902fc3e352ab3ebb64a2dbf41afc7dab5f77
-size 264242
+oid sha256:8844acf903ec50ff6390ef5cf264eb4433418c5bcd7a256635242a718636974d
+size 240647

results.json

@@ -1 +1 @@
-{"mean_reward": 162.5, "std_reward": 94.74307362546351, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-07-27T03:50:18.254345"}
+{"mean_reward": 319.0, "std_reward": 125.73384588089239, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-07-27T09:05:03.377842"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ac1a6cfdd9fb8e9e40f67e8692da1c60b777adb0e30ae7fd85af2556a94ff28
-size 28732
+oid sha256:922f003a28eed0e06c39318a20a57e5be0c163aad32b2552a8c7ca7ed60c2088
+size 43152