Error when running python -m omnigibson.examples.learning.navigation_policy_demo #957
-
Hey there, I just used Isaac Sim 4.1.0 and the new released OmniGibson v 1.1.1. I tried to run the demo of RL. However, there is an error: Could you help me to solve this issue? Best regards |
Beta Was this translation helpful? Give feedback.
Replies: 6 comments 3 replies
-
Could you let us know the version of |
Beta Was this translation helpful? Give feedback.
-
Hmm I am actually not able to reproduce your issue. The example ran okay on my side (also OG 1.1.1 and Isaac Sim 4.1). Maybe you can take a closer look at For your reference,
My |
Beta Was this translation helpful? Give feedback.
-
I downgrade my gymnasium version, now it works. Thanks a lot! |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
Another interesting phenomenon is that, when run a demo, after a while, the demo will terminate without any specific instruction, just a "Fatal Python error: Segmentation fault". Do you know how can I fix that? |
Beta Was this translation helpful? Give feedback.
I think the problem is that the neural network weights sit on one GPU and the observation you get from the environment (e.g. a RGB image tensor) sits on another GPU. You can try to make the policy model to sit on the same GPU as OmniGibson.