生成无向随机图

Question

I need help whenever i try to make the adj_matrix[u][v] == adj_matrix[v][u] for the same weight it doesn't apply and can anyone tell me if i am on the correct path ?

and is there a way to set the matrix without edge to an X instead of a 0?

  int gen_random_graph(int n)
 {
    srand(time(0));
    int adj_matrix[n][n];
    for(int u = 0; u < n; u++)
    {
        for (int v = u; v < n; v++)   //generating a N x N matrix  based on the # of vertex input
        {
            bool edgeOrNot = rand() % 2;   //decide whether it has an edge or not
            adj_matrix[u][v] = adj_matrix[v][u] = edgeOrNot;
            cout << u << " " << v << " " << adj_matrix[u][v] << endl;
            if(adj_matrix[u][v] == true)
            {
                adj_matrix[v][u] = true;
                if(u == v)                            //We can't have i = j in an undirected graph
                {
                    adj_matrix[u][v] = -1;
                }
                cout << u << " " << v << " " << adj_matrix[u][v] << endl;
            }
            else
            {
                adj_matrix[v][u] = -1;
                cout << u << " " << v << " " << adj_matrix[u][v] << "else" <<  endl;
            }
        }

    }

    for(int i = 0; i < n; i++)
    {
        for(int j = i; j < n; j++)           //create the N x N with edges and sets the weight between the edge randomly
        {
            if(adj_matrix[i][j] == true)
            {
                    int weight1 = rand()%10 + 1;
                    adj_matrix[i][j] = adj_matrix[j][i] = weight1;
                    cout << " ( " << i << "," << j << " ) " << "weight: " << adj_matrix[i][j] << endl;
            }
        }
    }
}
int main()
{
    int N;
    cout << "Enter number of vertices" << endl;
    cin >> N;
    gen_random_graph(N);

    return 0;
}

Answer 1

Instead 'X' I recommend use -1 value. There's a sample:

void gen_random_graph(int n) {
    srand(time(0));
    int adj_matrix[n][n];

    for(int u = 0; u < n; u++) {
        for (int v = u; v < n; v++) { //you don't need to calculate weight twice so loop starts from u
            if(v == u) {
                adj_matrix[u][v] = -1;
            }
            else {
                int weight = rand() % 10 - 1;
                adj_matrix[u][v] = adj_matrix[v][u] = weight;
            }
        }
    }
}

Now you can check value of edge. For -1 it doesn't exist.

Answer 2

I am not sure of your algorithm among other things. It's one thing to use int adj_matrix to store bool s, but storing 'X' , a char is something I would avoid.

For storing the same weights I would try:

const int weight = rand() % 10 + 1;
adj_matrix[u][v] = weight;
adj_matrix[v][u] = weight;

The compiler may try to invoke rand() twice, once for each statement. In the above you invoke rand() only once.

In general I would use STL containers such as std::vector<int> , and std::vector<std::vector<int>> , although the latter is not ideal from memory footprint point of view.

Instead of 'X' I would use std::numeric_limits<int>::min or max , which may not be equal to rand() % 10 + 1 .