Evolutionary-Optimization

A generic evolutionary algorithm for function optimization.

Introduction

This package allows the user to optimise a function using an evolutionary algorithm. An evolutionary algorithm uses the principles of evolution to find optimal solutions.

Using the Package

Getting Started

To get started with this package install this package:

pip install evolutionary-optimization-algorithm

Running Experiments

To run the code type the following in your terminal. The default experiment is a simple optimization of the $x^{2}$ using integers.

run_evolution

The parameters used for the run can be edited within the main.py file.

Personalising Experiments

To personalise your experiment you can either use the prebuilt phenotypes and genotypes using our interface, or you can build your own. To do so, you simply need to create a new phenotype / genotype class that inherits from the corresponding abstract class and implement the methods to suit your needs. You can mimic the structure of the main script to run your own experiments, like so:

    genotype_class = Genotype.get_genotype(Genotypes.<your_phenotype>)
    phenotype_class = Phenotype.get_phenotype(Phenotypes.<your_genotype>)
    fitness_function_class = FitnessFunction.get_fitness_function(FitnessFunctions.<your_fitness_function>)
    fitness_function_instance = fitness_function_class()

    evolutionary_algorithm = Evolution(
        phenotype=phenotype_class(genotype_class()),
        number_of_individuals=<desired_number_of_individuals>,
        number_of_generations=<desired_number_of_generations>,
        fitness_function=fitness_function_instance,
        ratio_of_elite_individuals=<desired_elitism_ratio>
    )

You can also plot your fitness over time and the phenotype over time:

    evolutionary_algorithm.plot_fitness_score_over_time()
    phenotype_function_points_tuple = generate_points_for_function(
        phenotype=evolutionary_algorithm.population.best_individual,
        bottom_plotting_limit=-10,
        upper_plotting_limit=10,
        number_of_points=100,
    )
    evolutionary_algorithm.plot_phenotype_function_and_best_individuals(phenotype_function_points_tuple)

or plot a gif of the algorithm over time:

    evolutionary_algorithm.create_gif(phenotype_function_points_tuple)