Installation

Requirements

Running studies? You don’t need Python.

This page is for using the scoring engine as a Python library. To run studies with the desktop instrument, download the installer instead — see the researcher quickstart.

The scoring engine targets Python 3.9+ with a deliberately small core (the engine is scipy-free so the frozen desktop sidecar stays lean):

Package	Minimum version	Used for
numpy	1.26	vectorized metric computation
pydantic	2.6	event and metric schema validation

Optional extras (pip install -e ".[scripts]") pull in matplotlib, pandas, and scipy for the figure/data scripts.

From source

git clone https://github.com/aslmylmz/open-bart.git
cd open-bart
pip install -e .

We recommend a virtual environment:

python -m venv .venv
source .venv/bin/activate        # Windows: .venv\Scripts\activate
pip install -e .

Verifying the install

The scoring engine is a plain Python package. From the repository root:

python -c "from scoring.bart import score_bart; print('ok')"

If that prints ok, you are ready to score a session.

Importing the package

The editable install (pip install -e .) registers the scoring package, so from scoring.bart import score_bart works from anywhere in the environment.

The game client

The React game client (app/src/BartGame.tsx) is a standalone component bundled by Vite into a static SPA (it is not published as an npm package). See Game client for integration details, including the backend endpoint it expects.

Building the documentation locally

pip install -r docs/requirements.txt
sphinx-build -b html docs docs/_build/html

Open docs/_build/html/index.html in a browser.