MATLAB Answers

A question about the simple genetic algorithm code

13 views (last 30 days)
Cantor Set
Cantor Set on 22 Aug 2019
Commented: Walter Roberson on 22 Aug 2019
This is a code I found of the genetic algorithm. I am a beginner in MATLAB but I really would like to understand the code. The CrossOver, Mutation and Select functions are written in separate m.files. I am reading the code linearly so it was all fine until I reached the line
Population=Initialize(NumberOfParameters, PopulationSize, LowerBounds, UpperBounds);
Is Initialize a saved function in MATLAB or is it a function the author is going to define later? I assumed it is goning to be defined later then when I proceed ..
What is getFitness, Population?! they were not defined before. I thought he is defining a new function named getFitness but then he did not define what Population is!
Here is the main code and the CrossOver, Mutation and Select are functions defined in seprate m.files.
%This code carries out a real Genetic Algorithm
PopulationSize=100; %number of chromosomes
NumberOfParameters=2; %number of designable parameters
NumberOfGenerations=5; %number of generations
LowerBounds=[-10 -10]'; %lower bounds on parameters
UpperBounds=[10 10]'; %lower bounds on parameters
MutationProbability=0.01; %probability of mutation
Fitness=zeros(PopulationSize,1); %fitness of all chromosomes
BestFitness=zeros(NumberOfGenerations,1); %storage for best fit in every iteration
BestFitnessPoints=zeros(NumberOfParameters,NumberOfGenerations); %storage for best fit
%point in every iteration
%now we do initialization
Population=Initialize(NumberOfParameters, PopulationSize, LowerBounds, UpperBounds);
%This is the main loop
for GenerationCounter=1:NumberOfGenerations %repeat for all generations
for i=1:PopulationSize %repeat for all chromosomes
Fitness(i,1)=getFitness(Population(:,i)); %get fitness of the ith element
[MaxValue MaxValueIndex]= max(Fitness); %get best fitness and its index
BestFitness(GenerationCounter,1)= MaxValue%store the best fit in this generation
BestFitnessPoints(:,GenerationCounter)=Population(:,MaxValueIndex); %store best fit point
%now we create the mating pool using the population
MatingPool=Select(Population,Fitness); %select the fittest from inside this population
%now we do simply crossover
OffSprings=CrossOver(MatingPool); %do simple crossover
Population=Mutation(OffSprings, MutationProbability); %do mutation
Any help will be very appreciated.
Thank you!


Walter Roberson
Walter Roberson on 22 Aug 2019
Population is defined by
Population=Initialize(NumberOfParameters, PopulationSize, LowerBounds, UpperBounds);
and then becomes input in the function call that is
Fitness(i,1)=getFitness(Population(:,i)); %get fitness of the ith element
Initialize is a function defined by the author.
Cantor Set
Cantor Set on 22 Aug 2019
Where is it defined? the author didn't write it in a separate m.fle
Thank you
Walter Roberson
Walter Roberson on 22 Aug 2019
I do not know. I am not a subscriber to scribd (which does not appear to be an official copy anyhow) and official copies of the book are over $CDN 100 in electronic format (though I could get a used hardcover version for $CDN 70 and more than a week shipping time)

Sign in to comment.

Answers (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!