Jetson Nano CUDA device not found

Autonomous Machines Jetson & Embedded Systems Jetson Nano
, ,
1

I’m using Ultralytics YOLOv8 trained model to make real-time predictions with OpenCV. When I run the script it says:

torch.cuda.is_available(): False
torch.cuda.device_count(): 0
os.environ[‘CUDA_VISIBLE_DEVICES’]: None
See Start Locally | PyTorch for up-to-date torch install instructions if no CUDA devices are seen by torch.

I tried export CUDA_HOME=/usr/local/cuda-10.0/
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64:$LD_LIBRARY_PATH

and I also tried the link given in the error, but nothing gets solved.

Can anyone help?

2

Can Any one please help???

4

Hi,

Please share how you install the PyTorch package.
Is the package built with CUDA support?

It’s recommended to try our PyTorch prebuilt below:
https://docs.nvidia.com/deeplearning/frameworks/install-pytorch-jetson-platform/index.html

More, for the latest JetPack 4.6.3, the CUDA version should be 10.2 rather than 10.0.

export LD_LIBRARY_PATH=/usr/local/cuda-10.2/lib64:$LD_LIBRARY_PATH

Thanks.

6

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.