twinify

What twinify is

twinify is an open-source package for creating synthetic twins of sensitive tabular datasets while applying differential privacy. It learns probabilistic models with ε and δ settings, where smaller values indicate stronger privacy. The package provides NAPSU-MQ and differentially private variational inference (DPVI). DPVI can automatically model categorical, continuous, or mixed data, or use models written with NumPyro. NAPSU-MQ is intended for fully categorical datasets. Automatic modeling can represent missing-value probabilities when missingness is independent across features. You can use twinify as a Python library or as a command-line tool that reads CSV files. Its implementation uses NumPyro, JAX, and d3p. A stable release can be installed from PyPI with pip, while a development version can come from a cloned repository. The code is licensed under Apache License 2.0 and is available for API or self-hosted use.