# py-why/dowhy

DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions. DoWhy is based on a unified language for causal inference, combining causal graphical models and potential outcomes frameworks.

Repository: https://github.com/py-why/dowhy
Canonical: https://ross.abutalabs.com/products/dowhy
Homepage: https://www.pywhy.org/dowhy
Language: Python
License: MIT
License Family: permissive
Topics: causal-inference, machine-learning, graphical-models, bayesian-networks, data-science, python3, causality, causal-models, treatment-effects, do-calculus, causal-machine-learning
Last push: 2026-08-26T13:55:19+00:00

## Health v2 (maintenance only)
Score: 76/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 32, longevity 100
- inputs: {"age_days": 3016, "days_push": 7, "days_rel": 298, "gap_med": 174.0, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8282, forks 1051 (observed 2026-08-28T04:10:20.064946+00:00)

## What it is
DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions. It unifies causal graphical models and potential outcomes frameworks to estimate causal effects and perform root-cause analysis.

## Use cases
- estimate the causal effect of a treatment on an outcome
- test causal assumptions before estimating effects
- perform root-cause analysis of latency spikes in microservices
- analyze causes of customer churn
- measure the impact of a loyalty program on revenue
- do causal attribution for changes in a supply chain

## When to choose
- you need principled causal effect estimation with explicit assumption modeling
- you want refutation tests and sensitivity analysis for causal estimates
- you need root-cause analysis using graphical causal models
- you prefer a Python ecosystem library that integrates with pandas and scikit-learn

## When to avoid
- you only need predictive machine learning without causal questions
- you need a point-and-click GUI for causal analysis
- your data has no plausible causal structure or assumptions to encode

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, data-science, analytics
- domain: data-science, machine-learning
- platform: python, cross-platform
- tags: causal-inference, causal-graphical-models, treatment-effects, do-calculus, bayesian-networks, root-cause-analysis, statistics

## Member repositories
- py-why/dowhy (main) score 76

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:20.064946+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 2026-08-29T17:28:06.589237+00:00, confidence not recorded.
  - readme: https://github.com/py-why/dowhy (fetched 2026-08-28T04:10:20.064946+00:00, sha 297396ba877d)
  - homepage: https://www.pywhy.org/dowhy (fetched 2026-08-29T08:27:48.943068+00:00, sha 1b72fa3a0eaa)
  - registry_pypi: https://pypi.org/pypi/dowhy/json (fetched 2026-08-29T08:27:48.951949+00:00, sha f180f92bfd90)
- Data as of 2026-08-30T08:39:29.467469+00:00.