Hai i need help in C++.. this is my code. By using this code I have to alter it and do Evolutionary Text/ Poem.
The poem is :
**“Berburu ke padang datar,
Dapat rusa berbelang kaki,
Berguru kepalang ajar,
Bagai bunga kembang tak jadi.” **
**There are 3 files.. Evo0.cpp(below code), mtrand.h and mtrand.cpp Click Here **
#include <iostream>
#include <ctime>
#include <windows.h>
#include <fstream>
#include "mtrand.h"
using namespace std;
int main () {
const int chromosome_length = 40;
const int pop_size = 20;
int max_generation = 200;
int mutate_position; // pos to mutate
int xover_position; // pos to xover
int rand_val; // to hold random value, i.e. allele
MTRand_int32 mt; // create MTRand object
mt.seed( time(NULL) ); // seed the RNG object
MTRand mtreal; // create MTRand object, real value of [0, 1) interval
mtreal.seed( time(NULL) ); // seed the RNG object
int pop[ pop_size ][ chromosome_length] = {0}; // un-init pop
int mating_pool[ pop_size ][ chromosome_length] = {0}; // un-init pop
int best[ chromosome_length ] = {0}; // un-init pop
int best_fitness = 0; // var to store best[] fitness value
int total_fitness = 0; // var to store total sum of fitness values
double avg_fitness = 0.0; // var to store avg fitness value
int fitness_values[ pop_size ]; // array to store fit value for all ind
ofstream myfile;
myfile.open ("output.txt");
// randomize pop here!----------------------------------------
for ( int i = 0; i < pop_size ; i++ ) {
cout << i << ": ";
for ( int j = 0; j < chromosome_length; j++ ) {
// get random value
//srand ( time(NULL) ); /* initialize random seed */
rand_val = 0; // re-init random value
rand_val = mt() % 2; /* generate random value - 0 or 1 */
pop[ i ][ j ] = rand_val;
//Sleep(1000);
cout << pop[ i ][ j ];
} // end for - inner loop
cout << endl;
} // end for - outer loop
cout << endl;
// randomize pop ENDS!----------------------------------------
int gen_num = 0; // init gen num
while ( gen_num < max_generation ) {
// re-init total & avg fitness values
total_fitness = 0;
avg_fitness = 0.0;
// output best ind here - use (highest) fitness value
int ind_fitness; // declare var for fitness
for ( int i = 0; i < pop_size; i++ ) {
ind_fitness = 0; // re-init fitness value
for ( int j = 0; j < chromosome_length; j++ ) {
ind_fitness = ind_fitness + pop[ i ][ j ];
} // end for - calc fitness value for one ind
fitness_values[ i ] = ind_fitness; // store every ind fitness
//compare current ind with (currently) best ind here
if ( ind_fitness > best_fitness ) {
for ( int k = 0; k < chromosome_length; k++ ) {
best[ k ] = pop[ i ][ k ];
} // end for loop - copy best ind
best_fitness = ind_fitness; // copy the current fitness into best ind
} // replace the best ind with current ind
// output (currently) best ind here - just to have a look!
cout << gen_num << ":" << i << " Current best ind: ";
// loop to print out best chromosome
for ( int m = 0; m < chromosome_length; m++ ) {
cout << best[ m ];
} // end for - print out best chromosome
cout << gen_num << ":" << i << " Current Best Fitness value: " << best_fitness << endl;
// summation of ind fitness values
total_fitness = total_fitness + fitness_values[ i ];
} // end for - calc fitness value for all ind
// avg fitness here!
avg_fitness = total_fitness / pop_size;
// dump data here!
myfile << gen_num << " " << best_fitness;
myfile << " " << avg_fitness << endl;
// parent selection----------------------------------------
int parent1;
int parent2;
int selected;
int z = 0;
int k = 2; // tournament selection parameter
// parent - tournament selection method
for ( int count = 0; count < pop_size; count++ ) {
parent1 = mt() % pop_size; /* generate random value, [0,pop_size) */
parent2 = mt() % pop_size; /* generate random value, [0,pop_size) */
if ( fitness_values[ parent1 ] > fitness_values[ parent2 ] ) {
selected = parent1;
} else {
selected = parent2;
}
for ( int m = 0; m < chromosome_length; m++ ) {
mating_pool[ z ][ m ] = pop[ selected ][ m ];
} // end for - copy winner into parent pool
z = z + 1; // increment parent pool counter
} // end for - loop each two individual
// print out parent pop
// parent selection ENDS----------------------------------
// xover here----------------------------------------
int p1[ chromosome_length ]; // temp parent1
int p2[ chromosome_length ]; // temp parent2
int offspring[ pop_size ][ chromosome_length ]; // offsprings array
int j = 0; // counter for index of 2nd parent
// prob of recombination
double Pc = 0.6; // arbitrary threshold value (higher than Pm)
double prob_recombine; // var to hold the generated prob
for ( int i = 0; i < pop_size; i = i + 2) {
j = i + 1;
for ( int k = 0; k < chromosome_length; k++ ) {
p1[ k ] = mating_pool[ i ][ k ];
p2[ k ] = mating_pool[ j ][ k ]; // p2 gets the next ind
}
//copy the chromosome array, i.e. the ind
for ( int z = 0; z < chromosome_length; z++ ) {
offspring[ i ][ z ] = p1[ z ];
offspring[ j ][ z ] = p2[ z ];
}
// get random position here for xover
// NO NEED to seed RNG object, already hv list of random numbers
//mt.seed( time(NULL) ); // seed the RNG object
xover_position = 0; // re-init pos value
// NEED to CHECK LOWER VALUE: should be
// 0 till (chromosome_length - 1)
xover_position = mt() % chromosome_length; /* generate xover position */
prob_recombine = mtreal();
// do recombination if prob LESS than threshold!
if ( prob_recombine < Pc ) {
for ( int counter = xover_position; counter < chromosome_length - 1; counter++ ) {
offspring[ i ][ counter ] = p2[ counter ];
offspring[ j ][ counter ] = p1[ counter ];
} // end for - loop for xover starting from xover_position
} // else here is just to display, actually no need!
else { // DO NOT do recombination if prob GREATER than threshold!
// maintain offspring i and j values as is!
cout << gen_num << ":" << i << " did not do recombination" << endl;
}
} // end for - loop select 2 parents
// xover ENDS----------------------------------------
// mutate here----------------------------------------
// prob of mutation
double Pm = 0.4; // arbitrary threshold value (lower than Pc)
double prob_mutate; // var to hold the generated prob
// go thru ALL ind
for ( int i = 0; i < pop_size; i++ ) {
// get random position here
// NO NEED to seed RNG object, already hv list of random numbers
//mt.seed( time(NULL) ); // seed the RNG object
mutate_position = 0; // re-init pos value
mutate_position = mt() % chromosome_length; /* generate position */
prob_mutate = mtreal();
// do mutation if prob LESS than threshold!
if ( prob_mutate < Pm ) {
// bit-flip the value at position
if ( offspring[ i ][ mutate_position ] == 0 ) {
offspring[ i ][ mutate_position ] = 1;
} else {
offspring[ i ][ mutate_position ] = 0;
} // end if-else bit-flip
} // else here is just to display, actually no need!
else { // DO NOT do mutation if prob GREATER than threshold!
cout << gen_num << ":" << i << " did not do mutation" << endl;
}
} // end for - mutate all ind
// mutate ENDS----------------------------------------
// survivor select here----------------------------------------
// since offspring is NOT in pop[][],
// so MUST write back into pop[][]
for ( int i = 0; i < pop_size; i++ ) {
for ( int j = 0; j < chromosome_length; j++ ) {
pop[ i ][ j ] = offspring[ i ][ j ];
} // end for - loop each ind
} // end for -loop write offspring into pop
// survivor select ENDS----------------------------------------
gen_num = gen_num + 1; // increment gen num
} // end while loop - GA finish
// output end best ind here
// output (the final) best ind here - just to have a look!
cout << "FINAL best ind: ";
// loop to print out best chromosome
for ( int j = 0; j < chromosome_length; j++ ) {
cout << best[ j ];
} // end for - print out best chromosome
cout << " FINAL Fitness value: " << best_fitness << endl;
myfile.close();
return 0;
} // end main
