mckinsey/causalnex
A Python library for causal inference and discovery using Bayesian networks, enabling data scientists to model cause-effect relationships.
Velocity · 7d
+1.0
★ / day
Trend
→steady
star history
CausalNex is a Python library developed by McKinsey’s QuantumBlack that helps data scientists infer causation rather than just observing correlation. It implements Bayesian network-based methods for causal discovery and inference, providing tools for structure learning, model fitting, and causal effect estimation. The library bridges traditional statistical causal analysis with modern machine learning approaches.