I’m looking for an example using the CUDA Toolkit in a Docker container with ubuntu on the Jetson TX2. Is this a supported combination? Does someone have a Dockerfile?
Hi,
There are two possible ways to do this.
1. You can use our l4t-base container below.
It will mount the CUDA toolkit from the Jetson natively to allow CUDA access within the docker.
https://catalog.ngc.nvidia.com/orgs/nvidia/containers/nvidia-l4t-base
2. You can also use our l4t-cuda image as a base image.
The container already has the CUDA toolkit installed.
Thanks.