Dear members,

I have implemented a Python library for modelling, inference and updating with�� Almost Desirable Gambles (ADG) models.�� It is both�� friendly and�� flexible, in the spirit of probabilistic programming.

It works with�� continuous, discrete and mixed variables.

Here you can find some additional info, setup instructions and 4 examples (notebooks):

The notebooks (and relative examples) are very simple, their purpose�� (at the moment) is only to highlight the functionalities of the library (for people who already know ADG).

I will soon add references to papers and books from the community and make them more tutorialish.

I welcome contributions,�� after all it is an open source project.

You can contribute in several ways:

1. giving me a feedback as user;

2. being an author of new notebooks where you can write down your favourite ADG models;

3. extending the library by including other functionalities, models etc (see https://github.com/PyRational/PyRational/ for instructions).

Alessandro Facchini and Dario Piga (IDSIA) will soon join the projects and help me to develop other models.

Have fun,

Alessio Benavoli