Reputation: 35
I am trying to raise a matrix to a power, using pointers but there is a mistake in my code I can't find.
#include <stdlib.h>
#include <stdio.h>
#include <conio.h>
>
int **alloc(int r, int c) {
int **d;
d = (int **)malloc(r * sizeof(int *));
for (int i = 0; i < r; i++) {
d[i] = (int *)malloc(c * sizeof(int));
}
return d;
void input(int **A, int r, int c) {
for (int i = 0; i < r; i++) {
for (int j = 0; j < c; j++) {
printf("[%d][%d]=", i, j);
scanf_s("%d", &A[i][j]);
}
}
}
void output(int **A, int r, int c) {
for (int i = 0; i < r; i++) {
for (int j = 0; j < c; j++) {
printf("%d ", A[i][j]);
}
printf("\n");
}
}
void power(int **A,int**D, int r, int c,int p) {
int i, j,k;
for (i = 0; i < r; i++) {
for (j = 0; j < c; j++) {
D[i][j] = A[i][j];
}
}
while (p) {
// this is the matrix multiplication, where I attempt to multiply my matrix A with itself, and store the result in D, which initially started as A's copy, and p is the power I'm raising it to.
for (i = 0; i < r; i++) {
for (j = 0; j < c; j++) {
for (k = 0; k < c; k++)
D[i][j] = D[i][j] + A[i][k] * D[k][j];
}
}
p--;
}
}
void main() {
int r, c;
int **A, **D;
printf("rows A: ");
scanf_s("%d", &r);
printf("columns A: ");
scanf_s("%d", &c);
A = alloc(r, c);
printf("\nValues of A:\n");
input(A, r, c);
printf("\nMatrIX A is:\n");
output(A, r, c);
D = alloc(r, c);
printf("input the value you want to raise your matrix to: ");
int p;
scanf_s("%d", &p);
power(A, D, r, c, p);
printf("\nMatrix raised to said power is:\n");
output(D, r, c);
_getch();
}
When I input the rows and columns as 2 each, and input all values in the matrix as 1 and raise it to the power of 3, my answer should be
4 4
4 4
instead of
72 72
232 232
What is wrong in my code? If I were to print the D matrix before the multiplication, it would print it correctly, as:
1 1
1 1
Upvotes: 1
Views: 178
Reputation: 29126
Your two-step allocation looks okay, except that you don't free the memory after you're done. Your problem is in how you multiply the matrix:
Raising a matrix to a power involves multiplying it to itself. You can only do that if the matrix is square. You can replace all occurrences of rows r
and columns c
with a single dimension n
.
When you do the actual multiplication:
D[i][j] = D[i][j] + A[i][k] * D[k][j];
you assign to D
and read from it at the same time. Subsequent calculations (of the same multiplication) will see a changed value of D[i][j]
. You will need a temporary "scratch" matrix.
Your code multiplies once too many. Also, Raising a matrix to the power of zero should yield the identity matrix.
Here's how your power
function could look like:
void power(int **A, int **D, int n, int p)
{
int i, j,k;
// assign identity matrix to result D
for (i = 0; i < n; i++) {
for (j = 0; j < n; j++) {
D[i][j] = (i == j);
}
}
// create a scratch matrix
int **tmp = alloc(n);
while (p-- > 0) {
// multiply [tmp] = [A] * [D]
for (i = 0; i < n; i++) {
for (j = 0; j < n; j++) {
tmp[i][j] = 0;
for (k = 0; k < n; k++)
tmp[i][j] += A[i][k] * D[k][j];
}
}
// copy [D] = [tmp]
for (i = 0; i < n; i++) {
for (j = 0; j < n; j++) {
D[i][j] = tmp[i][j];
}
}
}
// TODO: clean up the scratch matrix
}
Upvotes: 1