fgsl/geneticalgorithm

Fgsl Genetic Algorithm

1.0.0 2019-10-24 13:23 UTC

Requires

• php: >=7.0
• phpunit/phpunit: ^8.3

AGPL-3.0-only

• artificial intelligence
• computational intelligence
• evolutionary computation
• genetic algorithm

README

How to run the sample

There is a sample of genetic algorithm application as a an automated test. For running it, first run composer update for install the dependencies, and then run:

vendor/bin/phpunit --bootstrap vendor/autoload.php tests/

An example of test output:

...
Weight summatory:
1* 12 + 0 * 1 + 3 * 4 + 0 * 1 + 2 * 2 = 28

Value summatory:
1* 4 + 0 * 2 + 3 * 10 + 0 * 1 + 2 * 2 = 38

Weight summatory:
1* 12 + 2 * 1 + 0 * 4 + 1 * 1 + 0 * 2 = 15

Value summatory:
1* 4 + 2 * 2 + 0 * 10 + 1 * 1 + 0 * 2 = 9


Time: 8.62 seconds, Memory: 27.46 MB

OK (1 test, 1 assertion)

How to use:

Create a class that implements Fgsl\GeneticAlgorithm\RulesInterface. This class must define fitness function and other constraints.

Inject an instance of rules class in an instance of Fgsl\GeneticAlgorithm\GeneticAlgorithm and call method execute.

You can see processing details enabling verbose parameter in class constructor.