Invalid index to scalar variable: copying entries from one np array to another np array element-wise

Question

For context, I am writing code to compute the gray-level co-occurrence matrix in python for my data mining assignment.

when I have

    c = np.array([[1,1,2,1,3],
                  [2,1,2,3,3],
                  [1,2,1,1,3],
                  [1,3,1,2,1],
                  [3,3,2,1,1]])

and call glcm(c, 3, 2, 1) it matches with what the example we went over in class and no errors occur. I have the function call within the same cell as the function definition in Jupyter notebook. When I call the function on an image matrix (np array), I get an error saying invalid index to scalar variable on the line specified in the code below.

I find this weird because it works with the example (np-array named c) but does not work on other np arrays that represent grayscale images.

Am I populating C_shift incorrectly?

def glcm(C, k, mu, nu):
    C_rows = c.shape[0] 
    C_cols = c.shape[1]
    C_shift = np.zeros((C_rows, C_cols))
    C_shift[:] = np.nan # initialize everything to NaN

    # calculate C_shift
    for i in range(C_rows - mu):
        for j in range(C_cols - nu):
            C_shift[i][j] = C[i+mu][j+nu]  # ERROR: invalid index to scalar variable.

        
    # set the values of g
    g = np.zeros((k,k))
    for i in range(k):
        Ii = mat_map(C,i+1)
        for j in range(k):
            Ij = mat_map(C_shift, j+1)
            g[i][j] = np.multiply(Ii, Ij).sum()

    return g, g.sum()

Answer 1

EDIT:

I resolved it, turns out that I was not actually inputting a 2d array and was inputting a flattened array.

The array named 'c' above was what I thought I was inputting but I was actually inputting a flattened array (aka a vector). That is why when I try calling C[i+mu][j+nu] I was getting the invalid index to scalar variable because when C[i+mu] is called it would just return the i+mu'th number in the array. You cannot get the j+nu'th number of a scalar which is why the error occurred.