Evaluating and reproducing real-world robot manipulation policies (e.g., RT-1, RT-1-X, Octo) in simulation under common setups (e.g., Google Robot, WidowX+Bridge)
I run python simpler_env/simple_inference_visual_matching_prepackaged_envs.py --policy openvla/openvla-7b \
--ckpt-path None --task google_robot_move_near_v1 --logging-root ./results_simple_eval/ --n-trajs 10
I checked the troubleshooting section and followed the link's instructions,but it still didn't work.
Meet the error : RuntimeError: vk::Instance::enumeratePhysicalDevices: ErrorInitializationFailed
GPU : RTX3090
Ubuntu22.04
I run python simpler_env/simple_inference_visual_matching_prepackaged_envs.py --policy openvla/openvla-7b \ --ckpt-path None --task google_robot_move_near_v1 --logging-root ./results_simple_eval/ --n-trajs 10 I checked the troubleshooting section and followed the link's instructions,but it still didn't work. Meet the error : RuntimeError: vk::Instance::enumeratePhysicalDevices: ErrorInitializationFailed GPU : RTX3090 Ubuntu22.04