It seems we cannot even use A_in[i, :] ( or b[0, 1] ) in this exercise.
I know how to do it with the matrix multiplication method, but just wondering why it shows that “IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed” even if A_in is actually a 2D array.

It seems this is a bug?
Since in the cell below the coding cell(screenshot), both A_in and b are 2-D even if they are vectors, i.e. (1, n); however, in the test cell, A_in and b becomes 1-D (n, ), which is 1-D.