Open 1725917163 opened 10 months ago
Hi, could you provide more context on your use case? It seems like you want to perform reproducible (deterministic) experiments, you may want to look at the official guide on this topic: https://github.com/NVIDIA-Omniverse/OmniIsaacGymEnvs/blob/main/docs/reproducibility.md
Hello, we found a bug during model inference. When the headless parameter settings are different, the results of model inference are the same. The main reason is that the first frame state variable returned by the Isaac-sim simulator is inconsistent. In theory, it should are the same, is this normal?