Understand the project
Practical overview
When it fits
Use Dowhy when a data workflow needs an explicit causal model and treatment-effect analysis rather than predictive modelling alone. It is suited to examining the assumptions behind causal conclusions in Python-based projects.
Who it is for
- Python data scientists working with causal questions
- Machine-learning practitioners analysing treatment effects
- Researchers and engineers building causal-inference experiments
Documented uses
- Analyse treatment effects in observational datasets.
- Represent causal assumptions with graphical models.
- Combine graphical and potential-outcomes approaches in one analysis workflow.
- Test causal assumptions within machine-learning and data-science projects.
- Build reusable Python tooling for causal-inference experiments.
Before you choose
Requirements & limits
Dowhy is identified as open source under the MIT license and is available as a public GitHub repository. The technical profile lists desktop, server and self-hosted contexts.
- Python is the documented implementation language.
- The library combines causal graphical models with potential-outcomes methods.
- Explicit causal assumptions and assumption testing are central to its workflow.
- Its scope includes treatment effects, do-calculus and causal machine learning.
- The MIT license supports use within projects subject to its license terms.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Evidence and transparency
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 2 October 2026
- Original website title
- py-why/dowhy · Official homepage title · retained in its source language