Install NVIDIA DIGITS On Ubuntu 18.04

Ref Link:
https://github.com/dusty-nv/jetson-inference#system-setup

Installing the NVIDIA driver

Add the NVIDIA Developer repository and install the NVIDIA driver.

$ sudo apt-get install -y apt-transport-https curl build-essential
$ cat <<EOF | sudo tee /etc/apt/sources.list.d/cuda.list > /dev/null
deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 /
EOF
$ curl -s \
 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/7fa2af80.pub \
 | sudo apt-key add -
$ cat <<EOF | sudo tee /etc/apt/preferences.d/cuda > /dev/null
Package: *
Pin: origin developer.download.nvidia.com
Pin-Priority: 600
EOF
$ sudo apt-get update && sudo apt-get install -y --no-install-recommends cuda-drivers
$ sudo reboot

After reboot, check if you can run nvidia-smi and see if your GPU shows up.

$ nvidia-smi
Thu May 31 11:56:44 2018
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 390.30                 Driver Version: 390.30                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Quadro GV100        Off  | 00000000:01:00.0  On |                  Off |
| 29%   41C    P2    27W / 250W |   1968MiB / 32506MiB |     22%      Default |
+-------------------------------+----------------------+----------------------+

Installing Docker

Install prerequisites, install the GPG key, and add the Docker repository.

$ sudo apt-get install -y ca-certificates curl software-properties-common
$ curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
$ sudo add-apt-repository \
 "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"

Add the Docker Engine Utility (nvidia-docker2) repository, install nvidia-docker2, set up permissions to use Docker without sudo each time, and then reboot the system.

$ curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | \
  sudo apt-key add -
$ curl -s -L https://nvidia.github.io/nvidia-docker/ubuntu16.04/amd64/nvidia-docker.list | \
  sudo tee /etc/apt/sources.list.d/nvidia-docker.list
$ sudo apt-get update
$ sudo apt-get install -y nvidia-docker2
$ sudo usermod -aG docker $USER
$ sudo reboot

NGC Sign-up

Sign up to NGC if you have not.

https://ngc.nvidia.com/signup/register

Generate your API key, and save it somewhere safe. You will use this soon later.

Setting up data and job directory for DIGITS

Back on you PC (after reboot), log in to the NGC container registry

$ docker login nvcr.io

You will be prompted to enter Username and Password

Username: $oauthtoken
Password: <Your NGC API Key>

For a test, use CUDA container to see if the nvidia-smi shows your GPU.

docker run --runtime=nvidia --rm nvcr.io/nvidia/cuda:9.0-cudnn7-devel-ubuntu16.04 nvidia-smi

Setting up data and job directories

Create data and job directories on your host PC, to be mounted by DIGITS container.

$ mkdir /home/username/data
$ mkdir /home/username/digits-jobs

Starting DIGITS container

$ nvidia-docker run --name digits -d -p 8888:5000 \
 -v /home/username/data:/data:ro
 -v /home/username/digits-jobs:/workspace/jobs nvcr.io/nvidia/digits:18.05

Open up a web browser and access http://localhost:8888