https://github.com/tsoding/nnaskell

Neural Network in Haskell
https://github.com/tsoding/nnaskell

functional-programming hacktoberfest hacktoberfest2020 haskell machine-learning neural-network

Last synced: about 1 month ago
JSON representation

Neural Network in Haskell

• Host: GitHub
• URL: https://github.com/tsoding/nnaskell
• Owner: tsoding
• License: mit
• Created: 2017-10-31T19:36:40.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2019-08-03T15:00:59.000Z (about 6 years ago)
• Last Synced: 2025-04-29T12:55:28.551Z (5 months ago)
• Topics: functional-programming, hacktoberfest, hacktoberfest2020, haskell, machine-learning, neural-network
• Language: Haskell
• Size: 21.5 KB
• Stars: 7
• Watchers: 4
• Forks: 0
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • Changelog: ChangeLog.md
  • License: LICENSE

Awesome Lists containing this project

README

          
[![Tsoding](https://img.shields.io/badge/twitch.tv-tsoding-purple?logo=twitch&style=for-the-badge)](https://www.twitch.tv/tsoding)

[![Build Status](https://travis-ci.org/tsoding/nnaskell.svg?branch=master)](https://travis-ci.org/tsoding/nnaskell)

# Nnaskell

Neural Network in Haskell

## Quick Start

```console

$ nix-shell         # only for NixOS

$ stack build

$ stack ghci

> nn <- randomNN [2, 2, 1]

> xorTD

[([0.0,0.0],[0.0]),

 ([1.0,0.0],[1.0]),

 ([0.0,1.0],[1.0]),

 ([1.0,1.0],[0.0])]

> let trainedNN = head $ drop 1000 $ optimizeCost (cost xorTD) nn

> activateNN trainedNN $ vector [0.0, 0.0]

[4.045689665702377e-3]

> activateNN trainedNN $ vector [1.0, 0.0]

[0.9970477904605844]

>

```

## Support

You can support my work via

- Twitch channel: https://www.twitch.tv/subs/tsoding

- Patreon: https://www.patreon.com/tsoding