# mckinsey/causalnex

A Python library that helps data scientists to infer causation rather than observing correlation.

Repository: https://github.com/mckinsey/causalnex
Canonical: https://ross.abutalabs.com/products/causalnex
Homepage: http://causalnex.readthedocs.io/
Language: Python
License: NOASSERTION
License Family: other
Topics: causal-inference, causal-models, causal-networks, bayesian-networks, bayesian-inference, machine-learning, data-science, causalnex
Last push: 2026-08-10T14:48:10+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 8, longevity 100
- inputs: {"age_days": 2456, "days_push": 23, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2475, forks 291 (observed 2026-08-28T04:06:54.995749+00:00)

## What it is
CausalNex is a Python library for causal reasoning with Bayesian Networks, supporting structure learning, domain-expert augmentation of relationships, and estimation of intervention effects. It is now discontinued and archived by McKinsey as of June 2026, receiving no further updates or security patches.

## Use cases
- learn causal structure from data
- build bayesian networks for what-if analysis
- estimate effect of interventions on outcomes
- augment learned causal graphs with domain knowledge
- go beyond correlation to infer causation in python

## When to choose
- you need a mature archived codebase for bayesian network causal reasoning and accept no maintenance
- your project already depends on causalnex and you only need stability

## When to avoid
- you need actively maintained software with bug fixes and security patches
- starting a new project requiring long-term support

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, data-science
- domain: data-science, machine-learning
- platform: python
- tags: causal-inference, bayesian-networks, causal-discovery, what-if-analysis, discontinued, algorithms

## Member repositories
- mckinsey/causalnex (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:54.995749+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:28:23.176931+00:00, confidence not recorded.
  - readme: https://github.com/mckinsey/causalnex (fetched 2026-08-28T04:06:54.995749+00:00, sha d54c3a6c1881)
- Data as of 2026-08-30T08:39:29.467469+00:00.