Getting Started

Toward scalable, interactive detection of Simpson’s Paradox.

Published in:

To use the package, download (or clone) and:

cd DetectSimpsonParadox/
pip install .

To use the package, after installed:

import detect_simpsons_paradox as dsp

To use the data generation utilities, after installed:

import sp_data_utils as sp_data


To compile docs, sphinx and some extensions are required, all are included in requirements.txt and can be installed with

pip install -r requirements.txt


cd docs/
make html

To reinstall after changes

pip install --upgrade .

(only needed in development or after upgrade)

Project Organization

  • docs/source/ sphinx-based documentation for the project (can be md or rst)
  • docs/source/notebooks/ notebooks to be compiled in the documentation
  • notebooks/ a collection of research notes in jupyter notebooks related to the project
  • sp_data_util/ code that generates, downloads, and preprocesses data and generates synthetic data
  • detect_simpsons_paradox/ main module of tools for detecting Simpson’s paradox
  • data/ data for experiments, for reproducibility

Indices and tables