I finally released synloc 1.0.0. I had been meaning to do this since the paper on synloc was published in December 2025, but I kept postponing it; I finally got it done with the help of GPT 5.5.
This release makes the package feel properly stable: it adds better numeric-data validation, new quality and comparison tools such as compareStats, kolmogorov_distances, and quality_report, plus tests and documentation support. It also improves robustness in nearest-neighbor calculations, cluster sample allocation, and covariance estimation, especially in small-sample and edge-case settings.
Links
- Paper: Generating Synthetic Data with Locally Estimated Distributions for Disclosure Control
- Package repository: github.com/alfurka/synloc
- PyPI package: pypi.org/project/synloc
- Changelog: CHANGELOG.md
Related posts
If you want some background on synloc, I wrote about it before:
Have a nice day!