Matrix Multiplication in C

Hi,

I am trying to create a program in C that will multiply a matrix with a vector. Both the size of the vector and dimensions of the matrix are given by the user. I am having the most trouble trying to declare and use the 2D array, which is my matrix. I just don't get it. Here is what I have so far. there are probably a lot of errors because I don't quite understand how to use pointers correctly.

Code:

---

#include <stdio.h>
#include <stdlib.h>
int main(){
       
        int s, z;        /*s is number of columns, z number of rows and length of vector*/
        int i, x, y, ns, nz;
        double *X, *E, *a;                /*X is input vector, E is result vektor, a is matrix*/
        FILE *fd;
       
       
        /*Vektor*/
        fd = fopen("vektor3_2.dat", "r");
        fscanf(fd, "%d", &z);
        nz=z;
       
        if (z < 2) return 2;         
       
        X = malloc(z * sizeof(*X));   
        E = malloc(z * sizeof(*E));
        a = malloc(z * s * sizeof(*a));
       
        for (i = 0; i < z; i++)
                fscanf(fd, "%lf", &X[z]); 
       
       
       
        /*Matrix*/
        fscanf(fd, "%d", &s);
        ns=s;
        nz=z;
        for (z = 0; z<nz; z++)
                for (s=0; s<ns; s++)
                        fscanf(fd, "%lf", &a[z][s]);
               
       
        fclose(fd);
       
        /*Multiplication*/
       
        for (z=0; z<nz; z++)
                for (s=0; s<ns; s++)
                        E[z] = E[z] + a[z][s] * X[z];
       
        /*result out*/
        fd = fopen("Ergebnis3_2.dat", "w");
                for(i=0; i<z; i++)
                        printf("%lf\n", E[z]);
       
       
}

---

Code:

---

a = malloc(z * s * sizeof(*a));

---

Shouldn't you have some size for s at this point?

Code:

---

for (i = 0; i < z; i++)
                fscanf(fd, "%lf", &X[z]);

---

That should be i not z.


Quzah.

Ya, that was a dumb error. But as far as the matrix a goes, the compiler doesn't recognize it. This is the error:
pr3_2.c: In function ‘main’:
pr3_2.c:33: error: subscripted value is neither array nor pointer
pr3_2.c:42: error: subscripted value is neither array nor pointer

Code:

---

fscanf(fd, "%lf", &a[z][s]);

---

and

Code:

---

E[z] = E[z] + a[z][s] * X[z];

---

how declare the matrix and then use it?

> a = malloc(z * s * sizeof(*a));
Well you've allocated a flattened array (1D), not something you can subscript with [][]

If you want to sub-script by hand, then can do if you want.
fscanf(fd, "%lf", &a[z][s]);
would be
fscanf(fd, "%lf", &a[z*ns+s]);



Alternatively, you need

Code:

---

double **a;
a = malloc( z * sizeof(*a) );
for ( i = 0 ; i < z ; i++ ) a[i] = malloc( s * sizeof (*a[i]) );

---

This results in something you can use [][] notation on.

Ok, so in this new code, the first line creates a pointer to a. The second reserves locations for z (aka rows) and the third line reserves locations for columns within each row
Is that correct?

And so now I should be able to use [][] for each location in the matrix?

That fixed the compiling errors, but now I just get an empty matrix out (i.e 0 0 0 0). I can't tell if my error lies in my multiplication part, or the entry of data.

Dunno without seeing the latest code.

Code:

---

#include <stdio.h>
#include <stdlib.h>
int main(){
       
        int s, z;                        /*nummer von Spalten*/
        int i, x, y, ns, nz;
        double *X, *E, **a;                /*nummer von Zeilen*/
        FILE *fd;
       
       
        /*Vektor*/
        fd = fopen("vektor3_2.dat", "r");
        fscanf(fd, "%d", &z);
        nz=z;
       
        if (z < 2) return 2;          /* Programm beenden, wenn n < 2 */
       
        X = malloc(z * sizeof(*X));          /*Speicherplaetze fuer Vektorcomponenten reservieren*/
        E = malloc(z * sizeof(*E));
        a = malloc(z * sizeof(*a));
        for (i=0; i<z; i++)
                a[i] = malloc(s*sizeof(*a[i]));
       
        for (i = 0; i < z; i++)
                fscanf(fd, "%lf", &X[i]);  /*Vektor Inhalt bekommen*/
       
       
       
        /*Matrix*/
        fscanf(fd, "%d", &s);
        ns=s;
        nz=z;
        for (z = 0; z<nz; z++)
                for (s=0; s<ns; s++)
                        fscanf(fd, "%lf", &a[z][s]);
               
       
        fclose(fd);
       
        /*Multiplication*/
       
        for (z=0; z<nz; z++)
                for (s=0; s<ns; s++)
                        E[z] = E[z] + a[z][s] * X[z];
       
       
        fd = fopen("Ergebnis3_2.dat", "w");
                for(i=0; i<z; i++)
                        printf("%lf\n", E[z]);
       
       
}

---

Sorry about the comments in German.
s is number of columns
z is number of rows
E is result vector
X is input vector
a is input matrix

You still can't be using s in your malloc before you know what s is.

Ok, I moved my input of s above malloc.

Code:

---

#include <stdio.h>
#include <stdlib.h>
int main(){
       
        int s, z;                        /*nummer von Spalten*/
        int i, x, y, ns, nz;
        double *X, *E, **a;                /*nummer von Zeilen*/
        FILE *fd;
       
       
        /*Vektor*/
        fd = fopen("vektor3_2.dat", "r");
        fscanf(fd, "%d", &z);
        fscanf(fd, "%d", &s);
        ns=s;
        nz=z;
       
        if (z < 2) return 2;          /* Programm beenden, wenn n < 2 */
       
        X = malloc(z * sizeof(*X));          /*Speicherplaetze fuer Vektorcomponenten reservieren*/
        E = malloc(z * sizeof(*E));
        a = malloc(z * sizeof(*a));
        for (i=0; i<z; i++)
                a[i] = malloc(s*sizeof(*a[i]));
       
        for (i = 0; i < z; i++)
                fscanf(fd, "%lf", &X[i]);  /*Vektor Inhalt bekommen*/
       
       
       
        /*Matrix*/
       
        for (z = 0; z<nz; z++)
                for (s=0; s<ns; s++)
                        fscanf(fd, "%lf", &a[z][s]);
               
       
        fclose(fd);
       
        /*Multiplication*/
       
        for (z=0; z<nz; z++)
                for (s=0; s<ns; s++)
                        E[z] = E[z] + a[z][s] * X[z];
       
       
        fd = fopen("Ergebnis3_2.dat", "w");
                for(i=0; i<z; i++)
                        printf("%lf\n", E[z]);
       
       
}

---

I suppose if you want the right answer it might not hurt to set E[z] to 0 before starting to add things up.