Skip to main content

A sleek auto-differentiation library that wraps numpy.

Project description

mygrad is a simple, NumPy-centric autograd library. An autograd library enables you to automatically compute derivatives of mathematical functions. This library is designed to serve primarily as an education tool for learning about gradient-based machine learning; it is easy to install, has a readable and easily customizable code base, and provides a sleek interface that mimics NumPy. Furthermore, it leverages NumPy’s vectorization to achieve good performance despite the library’s simplicity.

This is not meant to be a competitor to libraries like PyTorch (which mygrad most closely resembles) or TensorFlow. Rather, it is meant to serve as a useful tool for students who are learning about training neural networks using back propagation.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

mygrad-1.4.1.tar.gz (85.6 kB view hashes)

Uploaded Source

Built Distribution

mygrad-1.4.1-py3-none-any.whl (88.0 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page