Project

Quantifying neural network interpretability: a benchmarking approach