PyCausalML 0.2.0

First GitHub release of PyCausalML. Classical estimators and the PyDeepCausalML 0.2.0 neural catalog install together as pycausalml 0.2.0.

Install

pip install https://github.com/zia207/PyCausalML/releases/download/v0.2.0/pycausalml-0.2.0-py3-none-any.whl

Requires Python ≥ 3.9. PyTorch ≥ 2.0 is included.

Added in 0.2.0

• bayesian_causalML: Bayesian quasi-experiments (difference-in-differences, regression discontinuity, interrupted time series, instrumental variables, synthetic control), Bayesian causal networks (backdoor and frontdoor identification, do queries), and Bayesian structural models for cross-sectional and time-series data.

Included from 0.1.0

• Tree and forest estimators, uplift learners, meta-learners, double machine learning, and doubly robust policy learners.
• Every public PyDeepCausalML estimator (TARNet, DragonNet, NOTEARS, TCDF, and the rest), importable from pycausalml and pydeepcausalml.
• Optional backends for DoWhy, EconML, causal-learn, Uber CausalML, DoubleML, CausalPy, and skgrf.

Documentation: https://zia207.github.io/PyCausalML/