Source code from our RecSys 2020 paper: "Making neural network interpretable with attribution: application to implicit signals prediction" (D. Afchar, R. Hennequin) - View it on GitHub