A Python library for Probabilistic Graphical Models

A Python wrapper for aGrUM, offering a user-friendly interface to create, model, learn, and calculate with Bayesian networks and other graphical models.

Get Started Install pyAgrum

Building a Bayesian network with pyAgrum

Comparing Bayesian networks with pyAgrum

Latest Releases

aGrUM/pyAgrum 2.3.2 released

January 26, 2026

pyAgrum
- Fix sklearn check_X_y compatibility with mixed-type DataFrames. ...

aGrUM/pyAgrum 2.3.1 released

December 11, 2025

aGrUM
- Fixed a typo in BIFXML export (thanks to Pierre-François Gimenez). ...

aGrUM/pyAgrum 2.3.0 released

October 27, 2025

aGrUM
- Use of std::format` (C++20) to correctly print doubles (thanks to Christopher Eveland). ...

Why pyAgrum?

Features	pyAgrum	pgmpy	bnlearn
Multi-PGM Support	✔	—	—
High-Performance C++ Core	✔	—	—
Causal Inference Tools	✔	✔	—
Influence Diagrams	✔	—	✔
Active Community	✔	✔	✔

Key Tasks

Classification

Build classifiers using Bayesian Networks for accurate predictions.

Learn More

Prediction

Make probabilistic predictions based on observed evidence.

Learn More

Causal Analysis

Discover causal relationships in your data with advanced algorithms.

Learn More

Decision Making

Optimize decisions under uncertainty with Influence Diagrams.

Learn More

Explanation

Explain model predictions using probabilistic reasoning.

Learn More

Model selection

Select the best model structure and parameters for your data.

Learn More

Ready to get started with pyAgrum?

Explore the documentation, join our community, and start building with pyAgrum today!

Install pyAgrum