#108108th of 168 in ML frameworks and librariesFor data and documents, data analysis
dowhy
Python library for causal effect estimation, refutation and root cause analysis
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 8.4k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Nov 2025
- Language
- Python
8,049 on 21 Apr 20268,401 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
DoWhy is a Python library for causal inference. It walks through modeling, identifying, estimating and refuting causal effects, and its graphical causal model module supports interventions, counterfactuals and anomaly attribution. It is not an LLM tool: it works on tabular data with NetworkX graphs and can call EconML estimators.
Who it is for: Data scientists doing causal inference and root cause analysis in Python
Strengths
- Four-step API (model, identify, estimate, refute) with refutation tests for any estimator
- Graphical causal model module covers counterfactuals, interventional sampling and root cause analysis
- Can use EconML CATE estimators such as DML through the same interface
- Many example notebooks and a user guide; installs via pip, poetry or conda
Weaknesses
- No LLM or agent features; unrelated to model serving or chat tooling
- Needs a causal graph or explicit causal assumptions from the user
- pygraphviz install can fail on some platforms
- Last release is 2025-11-08, so newer commits are not in a release
What it needs
- no GPU
- Needs Python 3.8+, NetworkX, pandas, numpy, scikit-learn, EconML (optional), pydot or pygraphviz (optional)
Quick answers
Is dowhy open source?
Yes. Its code is on GitHub under the MIT license.
Does dowhy need a GPU?
No, it runs without one.
Also in ML frameworks and libraries
See all 168| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |