GPU execution failed

[Phance]

[yolo_node-1] File "/home/hance/.local/lib/python3.10/site-packages/ultralytics/engine/predictor.py", line 306, in setup_model
[yolo_node-1] device=select_device(self.args.device, verbose=verbose),
[yolo_node-1] File "/home/hance/.local/lib/python3.10/site-packages/ultralytics/utils/torch_utils.py", line 192, in select_device
[yolo_node-1] raise ValueError(
[yolo_node-1] ValueError: Invalid CUDA 'device=gpu' requested. Use 'device=cpu' or pass valid CUDA device(s) if available, i.e. 'device=0' or 'device=0,1,2,3' for Multi-GPU.
[yolo_node-1]
[yolo_node-1] torch.cuda.is_available(): False
[yolo_node-1] torch.cuda.device_count(): 0
[yolo_node-1] os.environ['CUDA_VISIBLE_DEVICES']: None
[yolo_node-1] See https://pytorch.org/get-started/locally/ for up-to-date torch install instructions if no CUDA devices are seen by torch.
[yolo_node-1]
[ERROR] [yolo_node-1]: process has died [pid 5313, exit code 1, cmd '/home/hance/ros2_ws/install/yolo_ros/lib/yolo_ros/yolo_node --ros-args -r __node:=yolo_node -r __ns:=/yolo --params-file /tmp/launch_params_9dnamt_c -r image_raw:=/my_camera/pylon_ros2_camera_node/image_raw'].

@mgonzs13 I want to run YOLO detection using a GPU, but after installing CUDA, it still fails to run. What exactly should I do next? Is there anything else I need to install?

[mgonzs13]

@Phance, How have you installed CUDA? You have to install a nvidia driver compatible with your torch version, for example, nvidia-driver-550 and torch 2.4.0 in Ubuntu 22.

[Phance]

@mgonzs13 https://www.cnblogs.com/freedom-w/articles/17875258.html After I install this URL, it still shows an exception

[mgonzs13]

@Phance check the nvidia drive. You can use the nvidia-smi command.