C++ implements simple genetic algorithm
- 2020-04-02 03:07:25
- OfStack
This paper illustrates the C++ implementation of simple genetic algorithm. Share with you for your reference. The specific implementation method is as follows:
//Genetic algorithm GA
#include<iostream>
#include <cstdlib>
#include<bitset>
using namespace std;
const int L=5; //Define the length of the code
int f(int x) //Define the test function f(x).
{
int result;
result=x*x*x-60*x*x+900*x+100;
return result;
}
int main(int argc,char *argv[])
{
int a(0),b(32); //I'm going to define the domain of x
const int pop_size=8; //Define population size
// int L; // Specifies the length of the encoding
const int NG=20; //The algebra that specifies the maximum reproduction of a population
int t=0; //Algebra of current propagation
int p[pop_size]; //Define the population
int q[pop_size]; //Define a breeding population as the next generation of the population
srand(6553); //Defines the seed for random number generation
double sum; //That sum
double avl_sum; //Moderate average
double p_probability[pop_size]; //The probability of that
double pp[pop_size];
double pro; //Define the probability of random generation
float pc=0.90; //Define the probability of crossing
float pm=0.05; //Define the probability of variation
cout<<" Initial population ";
for(int i=0;i<pop_size;i++) //Generation 0 population
{
p[i]=rand()%31;
cout<<p[i]<<" ";
}
cout<<endl;
cout<<endl;
void Xover(int &,int &); //Declared cross function
//When the stop criterion is not satisfied, that is, the reproduction algebra does not reach the maximum algebra, the reproduction continues
while(t<=NG)
{
cout<<" Algebra of reproduction: t="<<t<<endl;
sum=0.0;
for(int i=0;i<pop_size;i++)
{
q[i]=p[i];
cout<<q[i]<<" ";
}
cout<<endl;
for(int i=0;i<pop_size;i++) //Calculate the sum
sum +=f(p[i]);
avl_sum=sum/pop_size;
cout<<"sum="<<sum<<endl;
cout<<" Moderate average ="<<avl_sum<<endl;
for(int i=0;i<pop_size;i++) // To calculate The probability of that
{
p_probability[i]=f(p[i])/sum;
if(i==0)
{
pp[i]=p_probability[i];
cout<<"pp"<<i<<"="<<pp[i]<<endl;
}
else
{
pp[i]=p_probability[i]+pp[i-1];
cout<<"pp"<<i<<"="<<pp[i]<<endl;
}
//cout<<"p_probability"<<i<<"="<<p_probability[i]<<endl;
}
//Choose parents
for(int i=0;i<pop_size;i++)
{
pro=rand()%1000/1000.0;
if(pro>=pp[0]&&pro<pp[1])
p[i]=q[0];
else if(pro>=pp[1]&&pro<pp[2])
p[i]=q[1];
else if(pro>=pp[2]&&pro<pp[3])
p[i]=q[2];
else if(pro>=pp[3]&&pro<pp[4])
p[i]=q[3];
else if(pro>=pp[4]&&pro<pp[5])
p[i]=q[4];
else
p[i]=q[5];
}
//Hybrid operator
int r=0;
int z=0;
for(int j=0;j<pop_size;j++)
{
pro=rand()%1000/1000.0;
if(pro<pc)
{
++z;
if(z%2==0)
Xover(p[r],p[j]);
else
r=j;
}
}
//Mutation operator
for(int i=1;i<=pop_size;i++)
for(int j=0;j<L;j++)
{
pro=rand()%1000/1000.0; //Generate random Numbers in the interval [0,1]
if(pro<pm)
{
bitset<L>v(p[i]);
v.flip(j);
p[i]=v.to_ulong();
}
}
t++;
cout<<endl; //Population generation
}
cout<<" Final results: ";
for(int i(0);i<pop_size;i++) //The algorithm ends, output the result
{
cout<<p[i]<<" ";
}
cout<<endl;
return 0;
}
//Define hybridization operation
void Xover(int &a,int &b)
{
int pos; //Randomly generated hybridization points are the number of components to exchange with each other
pos=rand()%5+1; //Of the n components, the pos component is randomly determined
int j,k;
j=pos;
k=pos;
bitset<L>e(a);
bitset<L>f(b); //The first pos components exchange with each other
bitset<L>g;
bitset<L>h;
for(int i=0;i<pos;i++)
{
if(e[i]==1)
g.set(i);
}
for(int i=0;i<pos;i++)
{
if(f[i]==1)
h.set(i);
}
for(j;j<L;j++)
{
if(f[j]==1)
g.set(j);
}
for(k;k<L;k++)
{
if(e[k]==1)
h.set(k);
}
a=g.to_ulong();
b=h.to_ulong();
}
Hope that the article described in the C++ programming to help you.