How to add zero vector or matrix to a 3D array in Python

Question

I have a 3D matrix:

array = np.array([[[1,2,3,4,5],[4,5,6,1,3], [7,8,9,6,7],[1,2,3,5,5],[4,5,6,8,9], [7,8,9,4,7]],[[1,2,3,4,5],[4,5,6,1,3], [7,8,9,6,7],[1,2,3,5,5],[4,5,6,8,9], [7,8,9,4,7]]])

This results the matrix as:

array([[[1, 2, 3, 4, 5],
        [4, 5, 6, 1, 3],
        [7, 8, 9, 6, 7],
        [1, 2, 3, 5, 5],
        [4, 5, 6, 8, 9],
        [7, 8, 9, 4, 7]],

       [[1, 2, 3, 4, 5],
        [4, 5, 6, 1, 3],
        [7, 8, 9, 6, 7],
        [1, 2, 3, 5, 5],
        [4, 5, 6, 8, 9],
        [7, 8, 9, 4, 7]]])

Now, I want to add a zero vector or matrix created using

x = np.zeros((1,5))

with the matrix array such that the result becomes:

array([[[1, 2, 3, 4, 5],
        [4, 5, 6, 1, 3],
        [7, 8, 9, 6, 7],
        [1, 2, 3, 5, 5],
        [4, 5, 6, 8, 9],
        [7, 8, 9, 4, 7],
        [0, 0, 0, 0, 0]],

      [[1, 2, 3, 4, 5],
        [4, 5, 6, 1, 3],
        [7, 8, 9, 6, 7],
        [1, 2, 3, 5, 5],
        [4, 5, 6, 8, 9],
        [7, 8, 9, 4, 7],
        [0, 0, 0, 0, 0]]])

Can anyone help? I tried with concatenation and stacks, including dstack but they didn't work (or I could not properly implement). I tried

a = np.vstack((array[0], x))
b =  np.vstack((array[1], x))

then I tried to join them as the 3D array using dstack as:

np.dstack((a, b))

this didn't work.

Direct concatenation (np.concatenate(array,x)) of course cannot work. That is why I tried to make a 3D zeros array but maybe np.zeros only create 2D arrays (or 1D given one dimension is 1)

Answer 1

Let's say arr is your (2, 6, 5) shaped array.

First make a suitably shaped array of zeros:

zeros = np.zeros((2, 1, 5))

This array has a 5-element row of zeros for each 'page' of your 3-D array.

Then use concatenate (since you want to join along an existing axis):

np.concatenate((arr, zeros), axis=1)

Result:

array([[[1., 2., 3., 4., 5.],
        [4., 5., 6., 1., 3.],
        [7., 8., 9., 6., 7.],
        [1., 2., 3., 5., 5.],
        [4., 5., 6., 8., 9.],
        [7., 8., 9., 4., 7.],
        [0., 0., 0., 0., 0.]],

       [[1., 2., 3., 4., 5.],
        [4., 5., 6., 1., 3.],
        [7., 8., 9., 6., 7.],
        [1., 2., 3., 5., 5.],
        [4., 5., 6., 8., 9.],
        [7., 8., 9., 4., 7.],
        [0., 0., 0., 0., 0.]]])

Answer 2

You can use np.append for this and then merge two 3d arrays using np.concatenate ie

a = np.array([[[1,2,3,4,5],[4,5,6,1,3], [7,8,9,6,7],[1,2,3,5,5],[4,5,6,8,9], [7,8,9,4,7]],[[1,2,3,4,5],[4,5,6,1,3], [7,8,9,6,7],[1,2,3,5,5],[4,5,6,8,9], [7,8,9,4,7]]])

z = np.zeros((1,5), dtype=int)
new = np.append(a[0], z, axis=0)
new1 = np.append(a[1], z, axis=0)

new2 = np.concatenate(([new], [new1]))

print(new2)

Output;

[[[1 2 3 4 5]
  [4 5 6 1 3]
  [7 8 9 6 7]
  [1 2 3 5 5]
  [4 5 6 8 9]
  [7 8 9 4 7]
  [0 0 0 0 0]]

 [[1 2 3 4 5]
  [4 5 6 1 3]
  [7 8 9 6 7]
  [1 2 3 5 5]
  [4 5 6 8 9]
  [7 8 9 4 7]
  [0 0 0 0 0]]]