Installing Compute Unified Device Architecture (CUDA)

If you do not have an Nvidia GPU or if you have an older Nvidia GPU that does not support CUDA, you can skip this step and move on to the next section, where we go over the PyTorch installation:

  1. Download the latest CUDA driver for your Nvidia GPU from the official Nvidia website here: https://developer.nvidia.com/cuda-downloads.
  2. Choose Linux under the operating system and your architecture (mostly x86_64), and then choose your Linux OS distribution (Ubuntu) version 14.04, 16.04, or 18.04, depending on your version, and select deb(local) as the installer type. That will download the cuda local installation file, named something like cuda-repo-ubuntu1604-8-0-local_8.0.44-1_amd64. Note your ...

Get Hands-On Intelligent Agents with OpenAI Gym now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.

Start your free trial