formulae 0.1.3

Formulas for mixed-effects models in Python

formulae

formulae is a Python library that implements Wilkinson's formulas for mixed-effects models. The main difference with other implementations like Patsy or formulaic is that formulae can work with formulas describing a model with both common and group specific effects (a.k.a. fixed and random effects, respectively).

This package has been written to make it easier to specify models with group effects in Bambi, a package that makes it easy to work with Bayesian GLMMs in Python, but it could be used independently as a backend for another library. The approach in this library is to extend classical statistical formulas in a similar way than in R package lme4.

Note: While this package is working, there is no online documentation yet and you may find bugs within the code. You are encouraged to play with this library and give feedback about it, but it is not recommended to incorporate formulae in a larger project at this early stage of development.

Installation

formulae requires a working Python interpreter (3.7+) and the libraries numpy, scipy and pandas with versions specified in the requirements.txt file.

Assuming a standard Python environment is installed on your machine (including pip), the latest release of formulae can be installed in one line using pip:

pip install formulae

Alternatively, if you want the development version of the package you can install from GitHub:

pip install git+https://github.com/bambinos/formulae.git

Documentation

The official documentation can be found here

Notes

• The data argument only accepts objects of class pandas.DataFrame.
• y ~ . is not implemented and won't be implemented in a first version. However, it is planned to be included in the future.