@liquid-carrot/carrot

<img src="https://raw.githubusercontent.com/liquidcarrot/carrot/master/images/carrot-logo_readme.png" alt="Carrot Logo" width="600px"/> <a href="https://travis-ci.org/liquidcarrot/carrot"> <img src="https://travis-ci.org/liquidcarrot/carrot.svg?branch=master" alt="Build Status via Travis CI"> </a> <a href="https://www.codacy.com/app/christianechevarria/carrot?utm_source=github.com&utm_medium=referral&utm_content=liquidcarrot/carrot&utm_campaign=Badge_Grade"> <img src="https://api.codacy.com/project/badge/Grade/3ee723b170f14b4990c8a0a6fc1feb27" alt="Codacy Badge"> </a> <a href="https://coveralls.io/github/liquidcarrot/carrot?branch=master"> <img src="https://coveralls.io/repos/github/liquidcarrot/carrot/badge.svg?branch=master" alt="Coverage Status"> </a> <a href="https://gitter.im/carrot-ai/community?utm_source=badge&utm_medium=badge&utm_campaign=pr-badgee&utm_content=badge"> <img src="https://badges.gitter.im/Join%20Chat.svg" alt="Join the chat at https://gitter.im/carrot-ai/community"> </a> <a href="/LICENSE"> <img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="Carrot's License"> </a> <a href="https://github.com/liquidcarrot/carrot/issues"> <img src="https://img.shields.io/badge/Made%20with%20%E2%99%A5%20by-Liquid%20Carrot-ff1414.svg" alt="Made with love"> </a> <h4>ℹ️ The new TypeScript version is coming! If you would like to try the expiremental version please clone the repository and checkout the typescript branch of the project. Docs for this new version can temporarily be found <a href="https://raimannma.github.io/carrot/index.html">here</a></h4> Carrot is an architecture-free neural network library built around neuroevolution Why / when should I use this? Whenever you have a problem that you: - Don't know how-to solve - Don't want to design a custom network for - Want to discover the ideal neural-network structure for You can use Carrot's ability to **design networks of arbitrary complexity by itself** to solve whatever problem you have. If you want to see Carrot designing a neural-network to play flappy-bird [check here](https://liquidcarrot.io/example.flappy-bird/) For Documentation, visit [here](https://liquidcarrot.github.io/carrot) ## Key Features - [Simple docs](https://liquidcarrot.github.io/carrot) & [interactive examples](https://liquidcarrot.io/example.flappy-bird/) - **Neuro-evolution** & population based training - Multi-threading & GPU (coming soon) - Preconfigured GRU, LSTM, NARX Networks - Mutable Neurons, Layers, Groups, and Networks - SVG Network Visualizations using D3.js ## Demos ![flappy bird neuro-evolution demo](https://raw.githubusercontent.com/liquidcarrot/carrot/master/images/flappy-bird-demo.gif) [Flappy bird neuro-evolution](https://liquidcarrot.io/example.flappy-bird/ "flappy bird playground") ## Install ```bash $ npm i @liquid-carrot/carrot ``` Carrot files are hosted by JSDelivr For prototyping or learning, use the latest version here: ```html <script src="https://cdn.jsdelivr.net/npm/@liquid-carrot/carrot/dist/carrot.umd2.min.js"></script> ``` For production, link to a specific version number to avoid unexpected breakage from newer versions: ```html <script src="https://cdn.jsdelivr.net/npm/@liquid-carrot/carrot@0.3.17/dist/carrot.umd2.min.js"></script> ``` ## Getting Started 💡 Want to be super knowledgeable about neuro-evolution in a few minutes? Check out [this article](https://www.oreilly.com/radar/neuroevolution-a-different-kind-of-deep-learning/ "Neuro-evolution based deep learning") by the creator of NEAT, Kenneth Stanley 💡 Curious about how neural-networks can understand speech and video? Check out [this video on Recurrent Neural Networks](https://www.youtube.com/watch?v=LHXXI4-IEns), from [@LearnedVector](https://github.com/LearnedVector), on YouTube This is a simple **perceptron**: ![perceptron](http://www.codeproject.com/KB/dotnet/predictor/network.jpg). How to build it with Carrot: ```javascript let { architect } = require('@liquid-carrot/carrot'); // The example Perceptron you see above with 4 inputs, 5 hidden, and 1 output neuron let simplePerceptron = new architect.Perceptron(4, 5, 1); ``` Building networks is easy with **6** built-in networks ```javascript let { architect } = require('@liquid-carrot/carrot'); let LSTM = new architect.LSTM(4, 5, 1); // Add as many hidden layers as needed let Perceptron = new architect.Perceptron(4, 5, 20, 5, 10, 1); ``` Building custom network architectures ```javascript let architect = require('@liquid-carrot/carrot').architect let Layer = require('@liquid-carrot/carrot').Layer let input = new Layer.Dense(1); let hidden1 = new Layer.LSTM(5); let hidden2 = new Layer.GRU(1); let output = new Layer.Dense(1); // connect however you want input.connect(hidden1); hidden1.connect(hidden2); hidden2.connect(output); let network = architect.Construct([input, hidden1, hidden2, output]); ``` Networks also shape **themselves** with neuro-evolution ```javascript let { Network, methods } = require('@liquid-carrot/carrot'); // this network learns the XOR gate (through neuro-evolution) async function execute () { // no hidden layers... var network = new Network(2,1); // XOR dataset var trainingSet = [ { input: [0,0], output: [0] }, { input: [0,1], output: [1] }, { input: [1,0], output: [1] }, { input: [1,1], output: [0] } ]; await network.evolve(trainingSet, { mutation: methods.mutation.FFW, equal: true, error: 0.05, elitism: 5, mutation_rate: 0.5 }); // and it works! network.activate([0,0]); // 0.2413 network.activate([0,1]); // 1.0000 network.activate([1,0]); // 0.7663 network.activate([1,1]); // 0.008 } execute(); ``` Build vanilla neural networks ```javascript let Network = require('@liquid-carrot/carrot').Network let network = new Network([2, 2, 1]) // Builds a neural network with 5 neurons: 2 + 2 + 1 ``` Or implement custom algorithms with neuron-level control ```javascript let Node = require('@liquid-carrot/carrot').Node let A = new Node() // neuron let B = new Node() // neuron A.connect(B) A.activate(0.5) console.log(B.activate()) ``` ## Try with #### Data Sets - [ ] [MNIST](https://www.npmjs.com/package/mnist) ## Contributors ✨ This project exists thanks to all the people who contribute. We can't do it without you! 🙇 Thanks goes to these wonderful people ([emoji key](https://allcontributors.org/docs/en/emoji-key)):    <table> <tr> <td align="center"><a href="https://liquidcarrot.io/"><img src="https://avatars3.githubusercontent.com/u/21148993?v=4" width="100px;" alt=""/> Luis Carbonell</a> <a href="https://github.com/liquidcarrot/carrot/commits?author=luiscarbonell" title="Code">💻</a> <a href="#ideas-luiscarbonell" title="Ideas, Planning, & Feedback">🤔</a> <a href="https://github.com/liquidcarrot/carrot/pulls?q=is%3Apr+reviewed-by%3Aluiscarbonell" title="Reviewed Pull Requests">👀</a> <a href="https://github.com/liquidcarrot/carrot/commits?author=luiscarbonell" title="Documentation">📖</a></td> <td align="center"><a href="http://liquidcarrot.io"><img src="https://avatars2.githubusercontent.com/u/23618650?v=4" width="100px;" alt=""/> Christian Echevarria</a> <a href="https://github.com/liquidcarrot/carrot/commits?author=christianechevarria" title="Code">💻</a> <a href="https://github.com/liquidcarrot/carrot/commits?author=christianechevarria" title="Documentation">📖</a> <a href="#infra-christianechevarria" title="Infrastructure (Hosting, Build-Tools, etc)">🚇</a></td> <td align="center"><a href="https://github.com/dan-ryan"><img src="https://avatars1.githubusercontent.com/u/775672?v=4" width="100px;" alt=""/> Daniel Ryan</a> <a href="https://github.com/liquidcarrot/carrot/issues?q=author%3Adan-ryan" title="Bug reports">🐛</a> <a href="https://github.com/liquidcarrot/carrot/pulls?q=is%3Apr+reviewed-by%3Adan-ryan" title="Reviewed Pull Requests">👀</a></td> <td align="center"><a href="https://github.com/IviieMtz"><img src="https://avatars0.githubusercontent.com/u/50965172?v=4" width="100px;" alt=""/> IviieMtz</a> <a href="https://github.com/liquidcarrot/carrot/commits?author=IviieMtz" title="Tests">⚠️</a></td> <td align="center"><a href="https://github.com/nicszerman"><img src="https://avatars3.githubusercontent.com/u/14032356?v=4" width="100px;" alt=""/> Nicholas Szerman</a> <a href="https://github.com/liquidcarrot/carrot/commits?author=nicszerman" title="Code">💻</a></td> <td align="center"><a href="http://www.threeceemedia.com"><img src="https://avatars3.githubusercontent.com/u/1046501?v=4" width="100px;" alt=""/> tracy collins</a> <a href="https://github.com/liquidcarrot/carrot/issues?q=author%3Atracycollins" title="Bug reports">🐛</a></td> <td align="center"><a href="https://github.com/raimannma"><img src="https://avatars3.githubusercontent.com/u/26171511?v=4" width="100px;" alt=""/> Manuel Raimann</a> <a href="https://github.com/liquidcarrot/carrot/issues?q=author%3Araimannma" title="Bug reports">🐛</a> <a href="https://github.com/liquidcarrot/carrot/commits?author=raimannma" title="Code">💻</a> <a href="#ideas-raimannma" title="Ideas, Planning, & Feedback">🤔</a></td> </tr> </table>    This project follows the [all-contributors](https://github.com/all-contributors/all-contributors) specification. Contributions of any kind welcome! ## 💬 Contributing [![Carrot's GitHub Issues](https://img.shields.io/github/issues/liquidcarrot/carrot.svg)](https://github.com/liquidcarrot/carrot/issues) Your contributions are always welcome! Please have a look at the [contribution guidelines](https://github.com/liquidcarrot/carrot/blob/master/CONTRIBUTING.md) first. 🎉 To build a community welcome to all, Carrot follows the [Contributor Covenant](https://github.com/liquidcarrot/carrot/blob/master/CODE_OF_CONDUCT.md) Code of Conduct. And finally, a big thank you to all of you for supporting! 🤗 <details><summary>Planned Features</summary> * [ ] Performance Enhancements * [ ] GPU Acceleration * [ ] Tests * [ ] Benchmarks * [ ] Matrix Multiplications * [ ] Tests * [ ] Benchmarks * [ ] Clustering | Multi-Threading * [ ] Tests * [ ] Benchmarks * [ ] Syntax Support * [ ] Callbacks * [ ] Promises * [ ] Streaming * [ ] Async/Await * [ ] Math Support * [ ] Big Numbers * [ ] Small Numbers </details> ## Patrons [![Carrot's Patrons](https://img.shields.io/endpoint.svg?color=blue&label=patrons&logo=patrons&url=https%3A%2F%2Fshieldsio-patreon.herokuapp.com%2Fliquidcarrot)](https://www.patreon.com/liquidcarrot) [![Become a Patron](https://c5.patreon.com/external/logo/become_a_patron_button.png)](https://www.patreon.com/liquidcarrot) ## Acknowledgements A special thanks to: [@wagenaartje](https://github.com/wagenaartje) for [Neataptic](https://github.com/wagenaartje/neataptic/) which was the starting point for this project [@cazala](https://github.com/cazala) for [Synaptic](https://github.com/cazala/synaptic/) which pioneered architecture free neural networks in javascript and was the starting point for Neataptic [@robertleeplummerjr](https://github.com/robertleeplummerjr) for [GPU.js](https://github.com/gpujs/gpu.js) which makes using GPU in JS easy and [Brain.js](https://github.com/BrainJS/brain.js) which has inspired Carrot's development