{"repo":"scikit-learn-contrib/MAPIE","free":true,"listed":false,"github":"https://github.com/scikit-learn-contrib/MAPIE","clone":"git clone https://github.com/scikit-learn-contrib/MAPIE.git","description":"A scikit-learn-compatible library for estimating prediction intervals and controlling risks, based on conformal predictions.","language":"Jupyter Notebook","stars":1579,"topics":["regression","confidence-intervals","data-science","python","sklearn","classification","conformal-prediction","risk-control"],"license":"BSD-3-Clause","category":"machine-learning","readme_excerpt":"MAPIE - Model Agnostic Prediction Interval Estimator An open-source Python library for quantifying uncertainties and controlling the risks of machine learning models. 🚀 MAPIE in 2026 🚀 New features have been implemented, starting with the application of risk control to emerging use cases such as LLM-as-Judge and image segmentation . In addition, exchangeability tests have been introduced to help users verify when MAPIE can be legitimately applied. Also, new adaptive conformal prediction methods have been added. Finally, the documentation has been updated with a new design! 🎉 MAPIE in 2025 🎉 MAPIE v1 is live! This new version introduces major changes to the API. Check out the release notes. See GitHub Releases and HISTORY.md for up-to-date details on the new features. Image credits: Cemrecan Yurtman (portrait) and hogrmahmood (zebra-horse hybrid). MAPIE allows you to: - Compute prediction intervals or prediction sets for regression, classification, and time series by estimating your model uncertainty on a conformalization dataset. - Control risks of more complex tasks such as multi-label classification and semantic segmentation in computer vision, providing probabilistic guarantees on metrics like recall and precision. - Easily use any model (scikit-learn, TensorFlow, PyTorch) thanks to scikit-learn-compatible wrapper if needed. MAPIE is part of the scikit-learn-contrib ecosystem. MAPIE relies notably on the fields of Conformal Prediction and Distribution-Free Inference. I","default_branch":null,"files":null,"tree":[],"storefront":"/r/scikit-learn-contrib","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/scikit-learn-contrib/MAPIE/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}