pushing model

README.md

@@ -16,7 +16,7 @@ model-index:
       type: Swimmer-v4
     metrics:
     - type: mean_reward
-      value: 105.46 +/- 3.12
+      value: 77.44 +/- 2.78
       name: mean_reward
       verified: false
 ---
@@ -46,7 +46,7 @@ curl -OL https://huggingface.co/sdpkjc/Swimmer-v4-ppo_fix_continuous_action-seed
 curl -OL https://huggingface.co/sdpkjc/Swimmer-v4-ppo_fix_continuous_action-seed5/raw/main/pyproject.toml
 curl -OL https://huggingface.co/sdpkjc/Swimmer-v4-ppo_fix_continuous_action-seed5/raw/main/poetry.lock
 poetry install --all-extras
-python ppo_fix_continuous_action.py --save-model --upload-model --hf-entity sdpkjc --env-id Swimmer-v4 --seed 5 --track
+python ppo_fix_continuous_action.py --save-model --upload-model --hf-entity sdpkjc --env-id Swimmer-v4 --seed 5 --track --cuda False
 ```
 
 # Hyperparameters
@@ -56,7 +56,7 @@ python ppo_fix_continuous_action.py --save-model --upload-model --hf-entity sdpk
  'capture_video': False,
  'clip_coef': 0.2,
  'clip_vloss': True,
- 'cuda': True,
+ 'cuda': False,
  'ent_coef': 0.0,
  'env_id': 'Swimmer-v4',
  'exp_name': 'ppo_fix_continuous_action',

events.out.tfevents.1704457861.4090-171.131299.0 → events.out.tfevents.1705723465.4090-171.2555519.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f46290dde322c60d133baf1894b5fe48c5322de3289e1a64bd319976460c811
-size 376390
+oid sha256:4175d2c14a00c0477ce120627fbe7ac3352b5377e268a7db8dbcd5f71a912478
+size 376391

ppo_fix_continuous_action.py

@@ -198,7 +198,7 @@ class NormalizeReward(gym.core.Wrapper, gym.utils.RecordConstructorArgs):
         return obs, rews, terminateds, truncateds, infos
 
     def reset(self, **kwargs):
-        self.returns = np.zeros(self.num_envs)
+        # self.returns = np.zeros(self.num_envs)
         return self.env.reset(**kwargs)
 
     def normalize(self, rews):