henrikbostrom/crepes

Python package for conformal prediction

66
/ 100
Established

Wraps standard scikit-learn classifiers and regressors to output calibrated p-values and prediction sets/intervals with guaranteed coverage through conformal prediction methods. Supports multiple strategies including standard, Mondrian, and class-conditional variants, with optional online calibration for semi-online prediction and exchangeability testing via martingale analysis. Integrates seamlessly with scikit-learn's ecosystem through wrapper classes while providing customizable non-conformity scores and difficulty estimates.

558 stars and 19,742 monthly downloads. Available on PyPI.

Maintenance 6 / 25
Adoption 20 / 25
Maturity 25 / 25
Community 15 / 25

How are scores calculated?

Stars

558

Forks

45

Language

Python

License

BSD-3-Clause

Last pushed

Oct 09, 2025

Monthly downloads

19,742

Commits (30d)

0

Dependencies

3

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/henrikbostrom/crepes"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.