Python package for Causal Discovery by learning the graphical structure of Bayesian networks. Structure Learning, Parameter Learning, Inferences, Sampling methods.

A Python 3 package for learning Bayesian Networks (DAGs) from data. Official implementation of the paper "DAGMA: Learning DAGs via M-matrices and a Log-Determinant Acyclicity Characterization"

Repository of a data modeling and analysis tool based on Bayesian networks

[Experimental] Global causal discovery algorithms

Official repository of the paper "Efficient Neural Causal Discovery without Acyclicity Constraints"

Automated Bayesian model discovery for time series data

Scalable open-source software to run, develop, and benchmark causal discovery algorithms

Amortized Inference for Causal Structure Learning, NeurIPS 2022

Graph Optimiser for Learning and Evolution of Models

DiBS: Differentiable Bayesian Structure Learning, NeurIPS 2021

Document Image Classification with Intra-Domain Transfer Learning and Stacked Generalization of Deep Convolutional Neural Networks

[AAAI 2020 Oral] Low-variance Black-box Gradient Estimates for the Plackett-Luce Distribution

Code associated with the paper "The World as a Graph: Improving El Niño Forecasting with Graph Neural Networks".

Sum-Product Network learning routines in python

Bayesian network structure learning

[SDM'23] ML4C: Seeing Causality Through Latent Vicinity

dagrad is a Python package that provides an extensible, modular platform for developing and experimenting with differentiable (gradient-based) structure learning methods.

The source code repository for the FactorBase system

Source code for the paper "Causal Modeling of Twitter Activity during COVID-19". Computation, 2020.

Experiments on structure learning of Bayesian Networks with emphasis on finding causal relationship

Python implementation of Bayesian Network Structure Learning using Quantum Annealing https://doi.org/10.1140/epjst/e2015-02349-9

Bayesian network analysis in R

Python implementation of "Characterizing Distribution Equivalence and Structure Learning for Cyclic and Acyclic Directed Graphs," in ICML 2020

Optimizing NOTEARS Objectives via Topological Swaps

Tractable learning of Bayesian networks from partially observed data

Standardizing Structural Causal Models, ICLR 2025

Code accompanying paper "Model-Augmented Conditional Mutual Information Estimation for Feature Selection" in UAI 2020

This is the official implementation of the bipartite matching experiment from the paper "Learning Randomly Perturbed Structured Predictors for Direct Loss Minimization".

OpenBNSL is a unified framework for fair, reproducible, and transparent comparison of Bayesian Network Structure Learning (BNSL) algorithms.

A curated list of causal structure learning research papers with implementations.