Projects and open-source libraries

A Python library for preparing tabular data for synthetic-data generation. It transforms real tabular data into Integer, Float, and Categorical (IFC) variables with NumPy-powered missing-value imputation, then applies the same learned inverse transformation to generated synthetic data so it can be mapped back to the original table structure. Supports automatic type inference, configurable fill strategies per column type, optional label encoding, and a portable fitted state that can be saved and reloaded across machines.

A Python library for generating synthetic tabular data using sequential tree models. It factorizes a table into a chain of conditional distributions — fitting a decision tree for each variable conditioned on all previous variables in the sequence — based on the approach of El Emam, Mosquera, and Zheng (2020). Supports greedy variable ordering, parallel fitting and sampling, LightGBM and scikit-learn tree backends, and within-leaf interpolation for continuous columns. Pure-Python core with no mandatory runtime dependencies.