Given a numeric input matrix A, possibly containing some zero values and some NaNs, replace any occurrences of zero or NaN with the character string 'error'. The output should be a cell array C of the same size as the input matrix. Each cell of the output cell array should either contain the corresponding entry of A if this is not zero or NaN, or the string 'error' otherwise.

Example:

If A = [1 0; NaN 1], then the output C should be the cell array C = {1, 'error'; 'error', 1}.