Getting Started

Installation

Via PyPI

The pyjpt package is available on the standard Python package index. Install the core package with

$ pip install pyjpt

Optional dependency groups can be installed depending on your use case:

$ pip install pyjpt[matplotlib]   # matplotlib and graphviz plotting
$ pip install pyjpt[plotly]       # interactive plotly plotting
$ pip install pyjpt[seq]          # sequential/temporal models
$ pip install pyjpt[mlflow]       # MLflow experiment tracking

Multiple groups can be combined:

$ pip install pyjpt[matplotlib,mlflow]

Via GitHub Repository

Alternatively, clone the repository and install from source:

$ git clone https://github.com/joint-probability-trees/jpt-dev
$ cd jpt-dev
$ pip install .

For an editable development install, see the Testing section below.

Build the Documentation

To build the documentation, clone the repository, switch to the doc folder and install the documentation requirements:

$ cd doc
$ pip install -r requirements.txt

After everything is successfully installed, the documentation can be built using Sphinx:

$ make html

After the build process has finished, you can view the documentation in your browser under build/html/index.html.

Supported Platforms

The package is tested on Ubuntu 22.04, but should work on other Linux distributions and macOS as well.

Testing

Running the test suite requires the dev dependency group, which includes all optional dependencies and test utilities:

$ pip install -e ".[dev]"

Run the full test suite using Python’s unittest discovery:

$ cd test
$ python -m unittest discover

To run a specific test file:

$ cd test
$ python -m unittest test_jpt