Fail to run NVIDIA Pytorch docker container in jetson nano card (Jetpack 4.2.1)

Robotics & Edge Computing Jetson & Embedded Systems Jetson Nano
, ,
1

I have pulled your docker image from this link, Nvidia PyTorch Container by

docker pull nvcr.io/nvidia/pytorch:20.10-py3

Using your command to run that docker image, and changing
local_dir:container_dir
to
/home/nano/examples/mnist:/data/mnist

  docker run --gpus all -it --rm -v /home/nano/examples/mnist:/data/mnist nvcr.io/nvidia/pytorch:20.10-py3

raised to me this error message:

docker: Error response from daemon: OCI runtime create failed: container_linux.go:349: starting container process caused “process_linux.go:449: container init caused "process_linux.go:432: running prestart hook 0 caused \"error running hook: exit status 1, stdout: , stderr: exec command: [/usr/bin/nvidia-container-cli --load-kmods configure --ldconfig=@/sbin/ldconfig.real --device=all --compute --utility --video --require=cuda>=9.0 --pid=27815 /var/lib/docker/overlay2/1b49f81ed097d3c3e9ba1958f7488fc06ef23774790ca2b539cb27bbc710a22c/merged]\\nnvidia-container-cli: mount error: mount operation failed: /usr/src/tensorrt: no such file or directory\\n\""”: unknown.

but tensorrt 5.1 has installed on jetson nano. Please help me to resolve this challenge.

2

Hi @barzanhayati, please use the l4t-pytorch container instead, which was built for Jetson. You will need JetPack 4.4 or newer on your Jetson to run the container. Example docker run commands used to start the container are found on the l4t-pytorch page.