This repository contains the code for experiments in LINA: a linearizing neural network architecture for accurate first-order and second-order interpretations.
If you have any question contact: adrien.f.badre-1@ou.edu
Each notebook contains its own function.
This was coded with python 3.6+, tensoflow 2.0+ and requires numpy and pandas.