Set up your machine¶

This page lists what you need to develop a model and evaluate it with Chap on your own machine. You do not need a Chap server, a database, the Modeling App or a separate chapkit install.

If the terminal, git or package managers are new to you, read Development tools first.

Windows: use WSL¶

Chap runs on macOS and Linux. On Windows, install WSL2 (Windows Subsystem for Linux) and run every command on these pages in the WSL terminal. See Chap on Windows.

1. Install uv¶

Chap is installed with uv, a tool for installing Python programs and packages. You need it even if your model is written in R.

macOS:

brew install uv

Linux and WSL:

curl -LsSf https://astral.sh/uv/install.sh | sh

Close and reopen the terminal, then check that it works:

uv --version

2. Install Chap¶

uv tool install chap-core --python 3.13

This installs the chap command. uv puts Chap and the packages it uses (more than a hundred) in a folder of their own, so the install does not change your other Python projects or your R setup. Check that it works:

chap --help

To update Chap later, run uv tool upgrade chap-core.

3. Install your model's language¶

R models: install R, and check that Rscript --version works in the terminal. RStudio is optional.
Python models: nothing more is needed. uv can install Python for your model.

4. Docker (only to deploy)¶

You do not need Docker to develop or evaluate a model. You need it when you package the model so that it can run next to a Chap installation. On macOS and Windows, install Docker Desktop. On Linux, install Docker Engine.