I was instructed by this site
to ask here so please forgive me and redirect me if needed.
I pulled and ran the following docker image
docker run --rm -p 10000:8888 -e JUPYTER_ENABLE_LAB=yes -v "$PWD":/workdir isbjornlabs/fastai-notebook-cuda9.2
Is the running container gpu enabled?
nvcc -V correctly identifies CUDA9.2
but when I issue ‘nvidia-smi’ from a terminal, it is an unknown command.
If I launch the same image with
nvidia-docker run --rm -p 10000:8888 -e JUPYTER_ENABLE_LAB=yes -v "$PWD":/workdir isbjornlabs/fastai-notebook-cuda9.2
the result correctly identifies my gpu.
MUST ALL images labeled as ‘gpu-enabled’ on docker hub be run with ‘nvidia-docker’ to use the gpu?