This is the avaiable code for the paper "An evidential classifier based on Dempster-Shafer theory and deep learning" (arXiv:2103.13549).
Codes for Dempster-Shafer layer and utility layer are in the file "libs", as well as the metrics "average utility".
The file "demo.ipynb" provides a demo about how to build, train, and interfere precise and imprecise classification with an evidential CNN classifier with the Dempster-Shafer layer and utility layers.
The file "weights_zoo" includes the parameters of a trained evidential CNN classifier that are used in the demo.
The required libraries and their version:
python == 3.7.10
tensorflow == 2.4.1.
#Update on Nov. 21th, 2021
Thanks to paul-bd for his/her reimplementation of Demspter-Shafer layer in the framework of Pytorch, see https://github.com/paul-bd/DempsterShaferTheory