Toolboxes & Code Repositories

Crepes-weighted

crepes-weighted extends crepes by adding support for weighted conformal prediction and predictive systems

software
Description:

crepes-weighted is an extension of crepes, a Python package that implements conformal classifiers, regressors, and predictive systems on top of any standard classifier and regressor. crepes-weighted extends crepes by adding support for weighted conformal prediction and predictive systems, in the future this could potentially be merged into the main crepes package.

More Information:

  • Originates from the AI-Driven Data Science research challenge (FAIR 2019–2023).

  • Responsible Research Lead: Sofie Van Hoecke

  • Authors:  Jef Jonke, Glenn Van Wallendael, Luc Duchateau,  Sofie Van Hoecke 

  • Programming Language: Python

  • Link to the repository

  • Categories: Supervised learning

Publications:

  • J. Jonkers, G. Van Wallendael, L. Duchateau, and S. Van Hoecke, “Conformal predictive systems under covariate shift,” in 13TH SYMPOSIUM ON CONFORMAL AND PROBABILISTIC PREDICTION WITH APPLICATIONS, Milan, Italy, 2024, vol. 230, pp. 406–423.