pushing model

README.md

@@ -16,7 +16,7 @@ model-index:
       type: Humanoid-v4
     metrics:
     - type: mean_reward
-      value: 572.69 +/- 142.36
+      value: 581.54 +/- 100.32
       name: mean_reward
       verified: false
 ---
@@ -46,7 +46,7 @@ curl -OL https://huggingface.co/sdpkjc/Humanoid-v4-ppo_fix_continuous_action-see
 curl -OL https://huggingface.co/sdpkjc/Humanoid-v4-ppo_fix_continuous_action-seed1/raw/main/pyproject.toml
 curl -OL https://huggingface.co/sdpkjc/Humanoid-v4-ppo_fix_continuous_action-seed1/raw/main/poetry.lock
 poetry install --all-extras
-python ppo_fix_continuous_action.py --save-model --hf-entity sdpkjc --seed 1 --upload-model --track --env-id Humanoid-v4
+python ppo_fix_continuous_action.py --save-model --upload-model --hf-entity sdpkjc --env-id Humanoid-v4 --seed 1 --track
 ```
 
 # Hyperparameters

events.out.tfevents.1700813989.4090-171.1716826.0 → events.out.tfevents.1704463091.4090-171.154748.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24c752cd3776a2b8351108716812cca8ebb32876fb484115f6112a9751e29178
+oid sha256:8ca1fa2a01c3986be509da03ea9c77799992cf9cf1cf744e2c92167ee7aa7de3
 size 1602161

ppo_fix_continuous_action.py

@@ -430,7 +430,7 @@ if __name__ == "__main__":
 
             # https://github.com/DLR-RM/stable-baselines3/pull/658
             for idx, trunc in enumerate(truncations):
-                if trunc:
+                if trunc and not terminations[idx]:
                     real_next_obs = infos["final_observation"][idx]
                     with torch.no_grad():
                         terminal_value = agent.get_value(torch.Tensor(real_next_obs).to(device)).reshape(1, -1)[0][0]