C - Genetic Algorithm for N Queens

I'm trying to figure our how to use the Genetic Algorithm to solve N queens.

Here is the program:

#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#define SIZE 8
#define POP 8

int answers[SIZE] = {5,3,1,7,4,6,0,2};

int getRand(int mod){
    if (mod==0) return 0;
    else return random()%mod;
}

void printArray(int array[]){
    int i;
    for(i=0; i<SIZE-1; i++) printf("(%i,%i),",i,array[i]);
    printf("(%i,%i)",SIZE-1,array[SIZE-1]);
    printf("\n");
}

int getWeight(int array[]){
    int weight = 28;
    int queen;
    for(queen=0;queen<SIZE;queen++){    //for each queen
        int nextqueen;
        for(nextqueen=queen+1;nextqueen<SIZE;nextqueen++){        //for each of the other queens (nextqueen = queen to avoid counting pairs twice)
            if(array[queen] == array[nextqueen] || abs(queen-nextqueen)==abs(array[queen]-array[nextqueen])){   //if conflict
                weight--;
            }
        }
    }
    return weight;
}

void geneticAlgorithm(){
    int population[POP][SIZE];
    int children[POP][SIZE];
    int weightProb[224] = {};
    int wpl = 0; //weightProb[] length
    float mutProb = 0.2; //higher prob yields faster times. works decently anyways. bug: prob = 0
    int done = 0;
    int i;
    for(i=0;i<POP;i++) for(int j=0;j<SIZE;j++) population[i][j] = getRand(SIZE);
    while(done == 0){        
        for(i=0;i<POP;i++){
            if(getWeight(children[i]) == 28){
                printf("solution: ");
                printArray(children[i]);
                done = 1;
            }
        }

        for(i=0;i<wpl;i++) weightProb[i] = (int)NULL; //clear weightprob
        wpl=0;

        //weighted probability distribution
        for(i=0;i<POP;i++){
            int w = getWeight(population[i]);
            for(int j=0;j<w;j++){
                weightProb[wpl] = i; //fill array with member number w times
                wpl++;
            }
        }

        //reproduce
        for(i=0;i<POP;i+=2){
            int par1 = weightProb[getRand(wpl)];
            int par2 = weightProb[getRand(wpl)];
            int split = getRand(SIZE);
            //crossover
            for(int j=0;j<split;j++){
                children[i][j] = population[par1][j];
                children[i+1][j] = population[par2][j];
            }
            for(int j=split;j<SIZE;j++){
                children[i][j] = population[par2][j];
                children[i+1][j] = population[par1][j];
            }
            //mutation
            if(getRand(1000000)<=mutProb*1000000){
                int child=getRand(2);
                if(child == 0) children[i][getRand(SIZE)] = getRand(SIZE);
                else children[i+1][getRand(SIZE)] = getRand(SIZE);
            }
        }
        for(i=0;i<POP;i++) for(int j=0;j<SIZE;j++) population[i][j] = children[i][j];
        wpl = 0;
    }
}

int main(int argc, const char * argv[]){
    srandom((unsigned int)time(NULL));  //seed random
    geneticAlgorithm();
    return 0;
}

The program runs and compiles properly but doesn't produce the results I'm after. I'm wanting to display the x,y co-ordinates of each queen and simply print them out on output. However instead I'm getting random garbage on output and I cant figure out why.

Current output example:

(0,0) (1,3248234234) (2,0) (3,-3248236736) (4,57435727) (5,234743567) (6, 23498348) (7,23487234)

Desired output example (a solution to the N queen problem):

(3,4) (7,2) (0,3) (4,6) (6,5) (1,7) (5,1) (2,0)

Not sure if this is the cause of your specific issue, but it is a problem:

for(i=0;i<POP;i++) for(int j=0;j<SIZE;j++) population[i][j] = getRand(SIZE);
while(done == 0){        
    for(i=0;i<POP;i++){
        if(getWeight(children[i]) == 28){
            printf("solution: ");
            printArray(children[i]);
            done = 1;
        }
    }

The above is the initialisation of the population and the beginning of the loop. On the first time through, the children array contains uninitialised data.

Given the rest of the algorithm, you need to be checking the weight of the elements of population (and printing an element from population) if it is a winner, not children.