About

Naive Automatic Differentiation Library (nadl) a small framework that can perform automatic differentiation. This framework is very simple implementation of how other big framworks do gradient calculation using Numerical Automatic Differentiation.

Installation

$ git clone https://github.com/siAyush/nadl.git
$ cd nadl
$ python3 setup.py install

Getting Started

The quickest way to start using nadl:

• Clone this repository and install.
• Make a python file and import nadl.
• Write your code.

Example:

from nadl.tensor import Tensor

t1 = Tensor([1, 2, 3], requires_grad=True)
t2 = Tensor([4, 5, 6], requires_grad=True)
t3 = t1 + t2

Testing

I have added a few basic unit tests. The command should be executed in the repository's root directory to run the test:

$ git clone https://github.com/siAyush/nadl.git
$ cd nadl
$ pytest