This repository contains the code used for the numerical experiments of the following paper:
U. Gazin, G. Blanchard, E. Roquain, "Transductive conformal inference with adaptive scores".
The code is under MIT Licence but please refer to and cite the above paper if you use it for academic purposes.
You need to download the four following packages (jdot.py, classif.py, procedure.py and algo.py) before running the code.
Files jdot.py, classif.py, procedure.py and algo.py are not from this paper.
jdot.py and classif .py can be found in the github page https://github.com/rflamary/JDOT/tree/master and is the implementation of the methods proposed by N. Courty, R. Flamary, A. Habrard, A. Rakotomamonjy, in "Joint Distribution Optimal Transportation for Domain Adaptation" published in Neural Information Processing Systems (NIPS), 2017.
procedure.py and algo.py can be found in the github page https://github.com/arianemarandon/adadetect#machine-learning-meets-fdr for the implementation and is the implementation of the methods proposed by Ariane Marandon, Lihua Lei, David Mary and Etienne Roquain in the paper "Adaptive novelty detection with false discovery rate guarantee", to appear in Annals of Statistics.