bnlearn
Python package for Causal Discovery by learning the graphical structure of Bayesian networks. Structure Learning, Parameter Learning, Inferences, Sampling metho
What is bnlearn?
Python package for Causal Discovery by learning the graphical structure of Bayesian networks. Structure Learning, Parameter Learning, Inferences, Sampling methods.
Overview of bnlearn
What it is
bnlearn is a Python package for causal discovery by learning the graphical structure of Bayesian networks. It provides pipelines for structure learning, parameter learning, inference, and sampling methods.
Key features
- Structure learning from data or with expert knowledge
- Parameter learning to estimate conditional probability distributions
- Causal inference for interventional and counterfactual distributions
- Generate synthetic data using Bayesian sampling
- Discretize continuous datasets for Bayesian network modelling
- Support for discrete, continuous, and mixed datasets
- Compare causal discovery libraries and benchmark LLMs against Bayesian methods
Use cases
bnlearn can be used for causal discovery, making causal predictions, exploring interventions, computing edge strength, imputing missing data, and comparing network structures. It is suitable for researchers and practitioners working with Bayesian networks and causal modelling.
Getting started
Python 3.10 or newer is required.Installation using pip:pip install bnlearnInstallation using uv:uv venv.venv/Scripts/activateuv pip install bnlearnInstallation from GitHub source:pip install git+https://github.com/erdogant/bnlearnAfter installation, import the library:import bnlearn as bn
License and pricing
The license is not stated in the provided metadata or README.
Based on the project's public README and repository metadata.