Skip to content

Releases

0.1.0

Status: released on 2026-09-30.

Version 0.1.0 is available from PyPI:

python -m pip install population-resemblance

The GitHub Release includes both the wheel and source distribution.

The initial public release establishes a tested implementation of categorical population resemblance monitoring with:

  • the Population Resemblance Statistic;
  • the full \(\delta\)-resemblance decision framework;
  • count-based and named-category APIs;
  • PSI and discrete KS benchmarks;
  • simulation and operating-characteristic tooling;
  • diagnostics and calibration sensitivity;
  • temporal monitoring;
  • reusable monitor objects;
  • serialization helpers;
  • an experimental independent two-sample API, explicitly outside the stable root API;
  • deterministic statistical property tests;
  • reproducible numerical benchmarks.

The complete change list is maintained in CHANGELOG.md.

Automated PyPI publishing

Publishing is triggered only when a GitHub Release is published.

The release workflow:

  1. checks that the release tag matches the package version;
  2. validates project metadata and the committed lockfile;
  3. builds both the source distribution and wheel;
  4. validates the built files with Twine;
  5. installs the built wheel in a clean virtual environment;
  6. publishes to PyPI using Trusted Publishing;
  7. attaches the same distribution files to the GitHub Release.

No PyPI password or API token is stored in the repository.

PyPI Trusted Publisher

PyPI publication uses Trusted Publishing with:

  • owner: DiogoRibeiro7;
  • repository: population-resemblance;
  • workflow: release.yml;
  • environment: pypi.

No repository secret containing a PyPI password or API token is required.

Release procedure

Update the package version first, merge all release preparation changes, then create and publish a GitHub Release with a matching tag such as v0.1.0.

If the tag and package version differ, the workflow stops before building or publishing.