2.5 GAMember.md

Introduction

GAMember is intended for those who wish not to use GP. Because a chromosome can be represented differently for different applications, this is only an abstract class that contains abstract methods to be implemented. If you wish to optimize a neural network using genetic algorithm, consider checking out my framework on neural networks LightNeuNet-MATLAB.

Properties

Name	Description
fitness	The fitness score of the individual.

Abstract Methods

Signature	Description
crossover(member, mateMember, opt)	Crossover operation in GA.
mutate(member, opt)	Mutation in GA.

crossover(member, mateMember, opt)

Replace part of the current individual's chromosome with the mateMember's.

Arguments

Argument	Description
mateMember	The other individual whose sections chromosome will be passed to the current member
opt	Options for developers to specify

mutate(member, opt)

Change part of the chromosome with randomly generated ones. The genetic operator introduces gene diversity into the population.

Arguments

Argument	Description
opt	Options for developers to specify

License

This project, including the documentation, is licensed under the MIT license. Copyright (c) 2022 Bohui WU.