genetics-js
v1.3.0
Published
Genetic and evolutionary algorithms framework for the web
Downloads
49
Readme
📚 Introduction
Evolutionary computing is one of the main techniques nowadays for solving complex optimization problems. This library provides with the basic structure for implementing the most common evolutionary algorithms, such as genetic algorithms.
Evolutionary algorithms basic structure
Evolutionary algorithms are composed basically by four elements:
- Individuals: Represent possible solutions of our problem in a determinate search space.
- Mutation: Mutation operator alterates one individual.
- Recombination: Recombination operator takes two parents and creates the offspring.
- Parent selection: Selection of the best parents that are going to be reproduced in the next generation.
- Survivor selection: Selection of the offspring and parents that are going to be the next generation.
This framework is going to provide the most common techniques for each component.
🔧 Installation
Currently project is under development (no stable version released :warning:), but it is going to be installed through npm:
npm install genetics-js
🧬 Usage
No major versions have been released, so only Individuals creation is implemented:
import Genetics from 'genetics-js';
const { BinaryIndividual } = Genetics.individual;
let individual = new BinaryIndividual('001100');
individual.genotype // [false, false, true, true, false, false]
🌠 Roadmap
The roadmap is strictly determined by the operations that are going to be implemented:
- [x]
v0.1.0
: Implementation of individuals. - [ ]
v0.2.0
: Implementation of mutation operators. - [ ]
v0.3.0
: Implementation of recombination operators. - [ ]
v0.4.0
: Implementation of parent selection methods. - [ ]
v0.5.0
: Implementation of survivor selection methods. - [ ]
v0.6.0
: Implementation of population and offspring management. - [ ]
v0.7.0
: Implementation of common evolutionary algorithms with fixed configurations.
🏆 Successes
The library has been successfully used in the following projects:
- Optimized planning of a trafficc light system using evolutionary algorithms - Francisco Arturo Cruz Zelante
👐 Contributing
You can report a bug, or request a feature with an issue:
Any help would be welcome :smile:.
💪 Authors
- Cristo Navarro - CristoNavarro - ([[email protected]]) - 2021
- Francisco Cruz - cruzelante098 - ([[email protected]]) - 2020
- Cristian Abrante - CristianAbrante - ([[email protected]]) - 2019
📝 License
This project is licensed under the MIT License