How to install RAPIDS from scratch on Ubuntu 18

Photo by David Dvořáček on Unsplash

Here is how to install RAPIDS in Ubuntu 18 from scratch

Install Nvidia Drivers

In this link, you can find how to install the Nvidia drivers on multiples OS

CUDA Toolkit 10.0 Archive

>> wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/cuda-ubuntu1804.pinsudo mv cuda-ubuntu1804.pin
>> sudo mv cuda-ubuntu1804.pin /etc/apt/preferences.d/cuda-repository-pin-600
>> wget http://developer.download.nvidia.com/compute/cuda/10.1/Prod/local_installers/cuda-repo-ubuntu1804-10-1-local-10.1.243-418.87.00_1.0-1_amd64.deb
>> sudo dpkg -i cuda-repo-ubuntu1804-10-1-local-10.1.243-418.87.00_1.0-1_amd64.deb
>> sudo apt-key add /var/cuda-repo-10-1-local-10.1.243-418.87.00/7fa2af80.pub
>> sudo apt-get update
>> sudo apt-get -y install cuda

Download conda from

Anaconda Python/R Distribution - Free Download

On CLI try

nvidia-smi

if you get the error

NVIDIA NVML Driver/library version mismatch

Maybe you need to restart the machine. Take a look a the link below if you need more info

NVIDIA NVML Driver/library version mismatch

sudo shutdown –r now

Install conda

From CLI execute(Maybe this could change depending on the file you download)

>> bash Anaconda3-2019.10-Linux-x86_64.sh

create conda environment

>> conda create --name rapids_blazing python=3.7
>> conda activate rapids_blazing

Install blazing and dependencies

>> conda install -c blazingsql/label/cuda10.0 -c blazingsql -c rapidsai -c nvidia -c conda-forge -c defaults blazingsql python=3.7 cudatoolkit=10.0

Install jupyter Notebook

>> conda install -c conda-forge notebook

Create a kernel

>> ipython kernel install --user --name=rapids_blazing

Run jupyter

jupyter notebook --ip=0.0.0.0 --port=8888

Go to your browser and point to your IP and the on port 8888