Skip to content

README.md

Latest commit

 

History

History
82 lines (59 loc) · 2.12 KB

README.md

File metadata and controls

82 lines (59 loc) · 2.12 KB

CUDA Development Container

This repo configures a CUDA enabled python development container. If you are unfamiliar with development containers, check out this page.

Installation

Choose one of the following ways to add this repo to your project.

Option A: Git submodule (recommended)

  1. Add repo as submodule
git submodule add git@github.com:raphaelschwinger/cuda-python-devcontainer.git .devcontainer

To update later:

git submodule update --remote .devcontainer

Option B: Download into .devcontainer

Use this if you do not want a submodule in your project.

One-time setup — clone only the files you need:

git clone --depth 1 git@github.com:raphaelschwinger/cuda-python-devcontainer.git .devcontainer

Or download a snapshot without git history:

mkdir -p .devcontainer
curl -L https://github.com/raphaelschwinger/cuda-python-devcontainer/archive/refs/heads/main.tar.gz \
  | tar xz --strip-components=1 -C .devcontainer

Add .devcontainer/ to your project's .gitignore if you do not want to track these files.

Updating from the main repo

If you used git clone, pull the latest changes:

cd .devcontainer
git pull
cd ..

If you used the tarball download, re-download and extract. Back up any local edits to devcontainer.json first (for example CUDA_VERSION and VARIANT):

mv .devcontainer/devcontainer.json /tmp/devcontainer.json.bak
rm -rf .devcontainer
mkdir -p .devcontainer
curl -L https://github.com/raphaelschwinger/cuda-python-devcontainer/archive/refs/heads/main.tar.gz \
  | tar xz --strip-components=1 -C .devcontainer
mv /tmp/devcontainer.json.bak .devcontainer/devcontainer.json

Configure and run

  1. Set CUDA_VERSION and VARIANT in .devcontainer/devcontainer.json to your desired values. The CUDA and Ubuntu version has to match with image tags found here.

  2. Reopen in container

  3. Init uv project

uv init
  1. Start installing packages
uv add <package>
  1. Run python in the virtualenv
uv run python