Undefined_AD Package

Undefined_AD Documentation

Brief Introduction

This project/package is part of the Harvard CS107/AC207 class final project in Fall 2021, automatic differentiation (AD). We created this package to compute the first-order derivative from a user-defined function automatically. Our package can utilize both forward and reverse modes. Additionally, we also allow users to define their own seeds vector for the Jacobian Matrix and the option to generate the computational graph associated with the input function.

To install, run the following command in your terminal

pip install undefined_AD

For details instruction on how to use this package, please follow the steps in the usage page.

Group Number:

Group 1

Group Members:

Xinran Tang xinran_tang@g.harvard.edu

Renhao Luo renhao_luo@hms.harvard.edu

Chelse Swoopes cswoopes@g.harvard.edu

Shijia Zhang shijiazhang@hms.harvard.edu

Harvard University, Fall 2021