Download the PHP package phpexperts/php-evolver without Composer
On this page you can find all versions of the php package phpexperts/php-evolver. It is possible to download/install these versions without Composer. Possible dependencies are resolved automatically.
Download phpexperts/php-evolver
More information about phpexperts/php-evolver
Files in phpexperts/php-evolver
Package php-evolver
Short Description A generic PHP Genetic Algorithm Framework.
License MIT
Homepage https://github.com/phpexpertsinc/php-evolver
Informations about the package php-evolver
PHP Evolver: A Genetic Algorithm Framework
Genetic Algorithms are a class of machine learning approaches that use the principles of natural selection, rather than the solving of mathematical formulae to find solutions to optimisation and search type problems. They are especially effective in complex situation that aren't easily "solved" and can often be used as a more-easily understood alternative to neural networks.
This framework takes care of most of the steps (loops) needed when developing and running a genetic algorithm, leaving you needing only to define the shape of your expected solution and a function to evaluate each candidate faciliating their comparison and thus the march towards an optimum.
Installation
You can install the package via composer:
Usage
Framing and Finding Solutions
Firstly create a class that defines a generic solution to the problem to be solved. The class must extend this package's Solution class, which will force the implemetation of two methods: genome() which defines the shape of a valid solution and evaluate(), which will calculate a numerical value that can be used to compare solutions.
Then instantiate and run the optimiser, creating an initial population of possible solutions to start its evaluation.
Data Manager
Although some use cases may not require much, if any, data against which to evaluate candidate solutions, others may need astronomical amounts. This could be be the case in financial markets where a trading strategy is sought and candidates are evaluated against the evolution of prices for many different securities, or in sports trading markets where possible strategies may be evaluated against changes in odds for thousands of events.
The DataManager class offers utilities optimised to assist with htese challenges. Here's the sort of thing that it can do:
PHP is also rather memory hungry when constructing arrays. If you experience out of memory errors, then the following may help:
Testing
Changelog
Please see CHANGELOG for more information on what has changed recently.
Contributing
Please see CONTRIBUTING for details.
Security
If you discover any security related issues, please email [email protected] instead of using the issue tracker. I take security very seriously and will welcome and respond promptly to your input.
Credits
Forked from https://github.com/ptercoles/genetic-algorithm-optimiser
- Theodore R. Smith
- Peter Coles
- All Contributors
License
The MIT License (MIT). Please see License File for more information.