pushing model

.gitattributes

@@ -35,3 +35,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 videos/Walker2d-v4__ppo_fix_continuous_action__4__1702930571-eval/rl-video-episode-8.mp4 filter=lfs diff=lfs merge=lfs -text
 replay.mp4 filter=lfs diff=lfs merge=lfs -text
+videos/Walker2d-v4__ppo_fix_continuous_action__4__1704457999-eval/rl-video-episode-8.mp4 filter=lfs diff=lfs merge=lfs -text
+videos/Walker2d-v4__ppo_fix_continuous_action__4__1704457999-eval/rl-video-episode-1.mp4 filter=lfs diff=lfs merge=lfs -text

README.md

@@ -16,7 +16,7 @@ model-index:
       type: Walker2d-v4
     metrics:
     - type: mean_reward
-      value: 3516.32 +/- 1737.55
+      value: 4387.63 +/- 1126.33
       name: mean_reward
       verified: false
 ---

events.out.tfevents.1702930579.4090-171.190027.0 → events.out.tfevents.1704458006.4090-171.133889.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e22d056aeac593dec4544cff456abdd72e37063546c3bcce43932fde8af905ea
-size 570817
+oid sha256:8e8b117e57d1c3e48c2516da5e7c6f6c7a5924d6196e03620ffde54d8bcaeb88
+size 570569

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 sha256:9791de4811dc3d72854742dabcd1923a3c2861faf6bcf7323c637f0743103864
-size 1439464
+oid sha256:eab52d9c0da8df2729246fcf075ffa47b283e13390acf8be04d1bafbd3f23d8c
+size 1456039

videos/Walker2d-v4__ppo_fix_continuous_action__4__1704457999-eval/rl-video-episode-1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab7b6e2867b857d199d2a718234c42d053232caeee09aa7d84c23cdbfbafdc8a
+size 1447338

videos/{Walker2d-v4__ppo_fix_continuous_action__4__1702930571-eval → Walker2d-v4__ppo_fix_continuous_action__4__1704457999-eval}/rl-video-episode-8.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9791de4811dc3d72854742dabcd1923a3c2861faf6bcf7323c637f0743103864
-size 1439464
+oid sha256:eab52d9c0da8df2729246fcf075ffa47b283e13390acf8be04d1bafbd3f23d8c
+size 1456039