Introduction to NumPy and Matplotlib > Reshape an array | Python Programming | Department of Computing

Introduction to NumPy and Matplotlib

Chapter 4: Array reshaping

Reshape an array

face Josiah Wang

Mmm… I’m digging that 3D array! Something is still not satisfying enough though!

Mission 4: Get the direction right!

Can you reshape the ( $3 \times 4$ ) np.array into a ( $2 \times 6$ ) np.array please? But this time, I want you to go by columns. So the first column should say [0, 4], the second column should say [-6, 1], the third column should be [5, -5], etc. I think they used to do that in MATLAB, so I kind of miss that kind of ordering!

x = np.array([[0, 1, 2, 3],
              [4, 5, 6, 7],
              [-6, -5, -4, -3]
             ])

reshaped_x = ????

assert np.all(reshaped_x == np.array([[ 0, -6,  5,  2, -4,  7],
                               [ 4,  1, -5,  6,  3, -3]]))

Hint: Check the possible input arguments for the .reshape() method.

By default np.arrays are reshaped in row-major order, so left-to-right, top-to-bottom for 2D. You can change this to column-major order (like in MATLAB), so top-to-bottom then left-to-right for 2D. This is done by passing the keyword argument order='F' (F stands for Fortran-like order).

reshaped_x = x.reshape((2, 6), order="F")