Metadata-Version: 2.2
Name: mikegrad
Version: 0.1.0
Summary: A minimal autograd engine
Home-page: https://github.com/canmike/mikegrad
Author: Can Michael Hucko
Author-email: c.michaelhucko@gmail.com
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=1.19.0
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# mikegrad

This repository is an implementation of Andrej Karpathy's ["Neural Networks: Zero to Hero - Micrograd"](https://www.youtube.com/watch?v=VMj-3S1tku0) video. The project implements backpropagation from scratch.

## Features

- Autograd engine for automatic differentiation
- Backpropagation implementation
- Basic neural network components
- Sigmoid activation function
- Example training with toy regression data

## TODO

- Refactor the codebase
- Publish to PyPI
- More activation functions
- More loss functions
- Support common optimizers (SGD, Adam)
- Add more examples
