v1.1.0

What's New 🤩

• Implement new binary risk control feature, see BinaryClassificationController and BinaryClassificationRisk. Here is a quick start
• See also the reworked risk control documentation
• MAPIE now supports Python versions up to the latest release (currently 3.13)

What's Changed 👀

• Revert incorrect renaming of calibration to conformalization in PrecisionRecallController
• Add scientific references for regression conformity scores
• Fix double inference when using predict_set function in split conformal classification
• Change prefit default value to True in split methods' docstrings to remain consistent with the implementation
• Fix issue 699 to replace TimeSeriesRegressor.partial_fit with TimeSeriesRegressor.update

Contributors 🙌🏻

Warm thanks to all contributors involved!

• @FaustinPulveric
• @gmartinonQM
• @Valentin-Laurent
• @vincentblot28
• @kleena
• @allglc

v1.0.1

What's Changed 👀

• Patch following v1.0.0 release: removing dependence to typing-extensions, making MAPIE unusable if this package is not installed

v1.0.0

🎉 MAPIE v1.0.0 is live! 🎉

• Major update, including a complete classification and regression public API rework, and a documentation revamp
• Other parts of the public API have been improved as well
• This update also includes bugfixes and developer experience improvements
• Here are extensive and user-focused release notes

Contributors 🙌🏻

Warm thanks to all contributors involved!

• [NEW] @C-BdB
• [NEW] @prncoprs
• [NEW] @sd29206
• [NEW] @FaustinPulveric
• [NEW] @Brijeshthummar02
• @sulphatet
• @gmartinonQM
• @Valentin-Laurent
• @jawadhussein462
• @vincentblot28

v1.0.0rc1

v1.0.0 coming soon 👀

v0.9.2

What's Changed 👀

• Fix issue 525 in contribution guidelines with syntax errors in hyperlinks and other formatting issues.
• Fix issue 495 to center correctly the prediction intervals
• Fix issue 528 to correct broken ENS image in the documentation
• Fix issue 548 to correct labels generated in tutorial
• Fix issue 547 to fix wrong warning
• Fix issue 480 (correct display of mathematical equations in generated notebooks)
• Temporary solution waiting for issue 588 to be fixed (optimize_beta not working)
• Remove several irrelevant user warnings
• Limit max sklearn version allowed at MAPIE installation
• Refactor MapieRegressor, EnsembleRegressor, and MapieQuantileRegressor, to prepare for the release of v1.0.0
• Documentation build: fix warnings, fix image generation, update sklearn version requirement
• Documentation test: add a doc testing step (in MAKEFILE and CI)
• Increase max line length from 79 to 88 characters
• Bump wheel version
• Other minor evolutions

Full Changelog: v0.9.1...v0.9.2

Contributors 🙌🏻

Warm thanks to all contributors involved!

• [NEW] @Leo-GG
• [NEW] @jawharmohammed
• [NEW] @sulphatet
• [NEW] @Valentin-Laurent
• @jawadhussein462
• @thibaultcordier
• @vincentblot28
• @LacombeLouis

v0.9.1

Fix of the issue #511 to keep the compatibility with older versions of MAPIE

v0.9.0

What's Changed

• 238 giving a fraction of samples instead of a number of samples in the subsample class by @BaptisteCalot in #464
• Change the sign in the equation of C_k in the Kolmogorov-Smirnov statistical test documentation by @BaptisteCalot in #470
• Unit tests for different subsamples by @BaptisteCalot in #468
• Building of Ensemble Classifier from MapieClassifier by @BaptisteCalot in #445
• ENH: externalize MapieClassifier methods into several non-conformity scores by @thibaultcordier in #476
• Update Github links that point to files in Mapie files by @BaptisteCalot in #486
• 473 replace assert nparray equal by nptestingassert array equal in all mapie unit tests by @BaptisteCalot in #485
• Update ts-changepoint-notebook by @BaptisteCalot in #487
• Import in the ts-changepoint notebook related to conformity scores is incorrect by @BaptisteCalot in #498
• Use the existing tutorial on ts changepoint to align the notebook by @BaptisteCalot in #490
• Add : **predict_params in fit and predict method for Mapie Regression by @BaptisteCalot in #471
• 493 add mondrian cp by @vincentblot28 in #504
• Add : **predict_params in fit and predict method for Mapie Classifier by @BaptisteCalot in #502
• chore: Refactor train-test split in plot_cqr_tutorial.py by @LacombeLouis in #482
• Refactor citation link in README.rst by @LacombeLouis in #509

Full Changelog: v0.8.6...v0.9.0

v0.8.6

What's Changed

• Updating sphinx version by @LacombeLouis in #466
• FIX bug #462: Decoupling infinite interval production, asymmetric conformal scores and quantile calculation by @thibaultcordier in #463

Full Changelog: v0.8.5...v0.8.6

v0.8.5

What's Changed

• FIX worklow issue with macos-python3.11 by @BaptisteCalot in #429
• chore: update miniconda version to latest stable by @LacombeLouis in #440
• FIX: Put lower threshold for difference by @LacombeLouis in #433
• FIX: change emails by @LacombeLouis in #436
• 425 readme page broken by @LacombeLouis in #435
• ENH: move get_true_label_position to utils in classification by @BaptisteCalot in #427
• chore: Update urllib to requests for downloading data by @LacombeLouis in #446
• 400 create citations using GitHub by @LacombeLouis in #442
• 249 clarification of the symmetry argument in cqr and more general documentation about cqr by @LacombeLouis in #443
• 415 documentation for winkler interval score by @LacombeLouis in #444
• Switch from the bostom dataset to the toy dataset by @thibaultcordier in #461
• FIX bug #452: Quantile Correction by @thibaultcordier in #453

Full Changelog: v0.8.3...v0.8.4

Full Changelog: v0.8.4...v0.8.5

v0.8.4

What's Changed

• See the v0.8.5 which replaces the v0.8.4

Full Changelog: v0.8.3...v0.8.4