Getting Started

The RAPIDS data science framework includes a collection of libraries for executing end-to-end data science pipelines completely in the GPU. It is designed to have a familiar look and feel to data scientists working in Python. Here’s a code snippet where we read in a CSV file and output some descriptive statistics:

import cudf
gdf = cudf.read_csv('path/to/file.csv')
for column in gdf.columns:

Try Now Online

Jump right into a GPU powered RAPIDS notebook.
Try with Colaboratory

Deploy RAPIDS on the Cloud

Learn how to deploy RAPIDS on Cloud Service Providers

GPU Powered Data Science

RAPIDS uses optimized NVIDIA CUDA® primitives and high-bandwidth GPU memory to accelerate data preparation and machine learning. The goal of RAPIDS is not only to accelerate the individual parts of the typical data science workflow, but to accelerate the complete end-to-end workflow.

We suggest that you take a look at the sample workflow in our Docker container (described below), which illustrates just how straightforward a basic XGBoost model training and testing workflow looks in RAPIDS.


GPU: NVIDIA Pascal™ or better with compute capability 6.0+

OS: Ubuntu 18.04/20.04 or CentOS 7/8 with gcc/++ 9.0+

  • See RDN 8 for recent changes to gcc/++ 9.0 requirements
  • RHEL 7/8 support is provided through CentOS 7/8 builds/installs

Docker: Docker CE v19.03+ and nvidia-container-toolkit

CUDA & NVIDIA Drivers: One of the following supported versions:

RAPIDS Release Selector

RAPIDS is available as conda packages, docker images, and from source builds. Use the tool below to select your preferred method, packages, and environment to install RAPIDS. Certain combinations may not be possible and are dimmed automatically. Be sure you’ve met the required prerequisites above and see the details below.

Conda  Preferred Install Method
Docker + Examples  Preferred Install Method
Docker + Dev Env  Advanced Install Method
Source  Advanced Install Method
Stable (21.10)
Nightly (21.12a)
RAPIDS and BlazingSQL
All Packages
Ubuntu 18.04 
Ubuntu 20.04 
CentOS 7 
CentOS 8 
RHEL 7&8 
Python 3.7
Python 3.8
CUDA 11.0
CUDA 11.2
# Javascript is needed for this tool to run, please make sure it is enabled

details below

Conda Install

You can get a minimal conda installation with Miniconda or get the full installation with Anaconda.

For instructions on how to build a development conda environment, see the cuDF README for more information. Also refer to the cuML README for conda install instructions for cuML.

Build From Source

Checkout the cuDF README, cuML README, or cuGraph README for from-source build instructions.

Where is Pip?

Refer to our RAPIDS 0.7 Release Drops PIP Packages — and sticks with Conda post for details on why we no longer support PIP installs.

Docker Container

The copied Docker command above should auto-run a notebook server. If it does not, run the following command within the Docker container to launch the notebook server.

bash /rapids/utils/

NOTE: This will run JupyterLab on your host machine at port 8888.

Legacy Docker Users

Docker CE v18 & nvidia-docker2 users will need to replace the following for compatibility:

'docker run --gpus all' with 'docker run --runtime=nvidia'

Use JupyterLab to Explore the Notebooks

Notebooks can be found in notebooks directory within the container:

/rapids/notebooks/cuml (cuML demos)

/rapids/notebooks/cugraph (cuGraph demos)

/rapids/notebooks/xgboost (XGBoost demo)

Advanced Usage

See the RAPIDS Container README page for more information about using custom datasets. Docker Hub and NVIDIA GPU Cloud host RAPIDS containers with full list of available tags.