pushing model

.gitattributes

@@ -37,3 +37,6 @@ videos/HalfCheetah-v4__ppo_fix_continuous_action__1__1700810944-eval/rl-video-ep
 videos/HalfCheetah-v4__ppo_fix_continuous_action__1__1700810944-eval/rl-video-episode-1.mp4 filter=lfs diff=lfs merge=lfs -text
 videos/HalfCheetah-v4__ppo_fix_continuous_action__1__1700810944-eval/rl-video-episode-0.mp4 filter=lfs diff=lfs merge=lfs -text
 replay.mp4 filter=lfs diff=lfs merge=lfs -text
+videos/HalfCheetah-v4__ppo_fix_continuous_action__1__1704463403-eval/rl-video-episode-8.mp4 filter=lfs diff=lfs merge=lfs -text
+videos/HalfCheetah-v4__ppo_fix_continuous_action__1__1704463403-eval/rl-video-episode-1.mp4 filter=lfs diff=lfs merge=lfs -text
+videos/HalfCheetah-v4__ppo_fix_continuous_action__1__1704463403-eval/rl-video-episode-0.mp4 filter=lfs diff=lfs merge=lfs -text

README.md

@@ -16,7 +16,7 @@ model-index:
       type: HalfCheetah-v4
     metrics:
     - type: mean_reward
-      value: 1791.89 +/- 50.82
+      value: 1777.06 +/- 25.88
       name: mean_reward
       verified: false
 ---
@@ -46,7 +46,7 @@ curl -OL https://huggingface.co/sdpkjc/HalfCheetah-v4-ppo_fix_continuous_action-
 curl -OL https://huggingface.co/sdpkjc/HalfCheetah-v4-ppo_fix_continuous_action-seed1/raw/main/pyproject.toml
 curl -OL https://huggingface.co/sdpkjc/HalfCheetah-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
+python ppo_fix_continuous_action.py --save-model --upload-model --hf-entity sdpkjc --env-id HalfCheetah-v4 --seed 1 --track
 ```
 
 # Hyperparameters

events.out.tfevents.1700810954.4090-171.1713824.0 → events.out.tfevents.1704463412.4090-171.159960.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a6094de6ce453fb7197d512837c5027130fd153522d4498e0a8a77908f8e1fd
+oid sha256:7fbb325da441b95f18efe776d8483af5329fbf53aa5c3e247edbada55754f81e
 size 376394

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]

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f388ab37e6cf7ffe716cb240e68a8c58d58cb8df2deed8599fd6c3be844aa730
-size 1924637
+oid sha256:6282fa1675fd6f69ed6fdf5e21d3b3e1c6ce023c0a77ad82162b4409a94648a7
+size 1947876

videos/{HalfCheetah-v4__ppo_fix_continuous_action__1__1700810944-eval → HalfCheetah-v4__ppo_fix_continuous_action__1__1704463403-eval}/rl-video-episode-0.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08f88b1a017886e7e5fd528d5e1f866b26e45825680ba80843f77d92faedb11a
-size 1931669
+oid sha256:3ace17e6c1cb3ced131fc1d7f6a80beefec2542a775acf4e139041d2688caafa
+size 1972380

videos/{HalfCheetah-v4__ppo_fix_continuous_action__1__1700810944-eval → HalfCheetah-v4__ppo_fix_continuous_action__1__1704463403-eval}/rl-video-episode-1.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7895e868ea17b8f6cbcec4b6ef7c82326997b438512e2df5c0761261312d337d
-size 1985925
+oid sha256:a219e9231e637e722eca508385765445e3da213594d79010272f9532fd9afaaf
+size 1988387

videos/{HalfCheetah-v4__ppo_fix_continuous_action__1__1700810944-eval → HalfCheetah-v4__ppo_fix_continuous_action__1__1704463403-eval}/rl-video-episode-8.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f388ab37e6cf7ffe716cb240e68a8c58d58cb8df2deed8599fd6c3be844aa730
-size 1924637
+oid sha256:6282fa1675fd6f69ed6fdf5e21d3b3e1c6ce023c0a77ad82162b4409a94648a7
+size 1947876