Creating rank 3 numpy arrays

Question

I would like to create a rank 3 array, using numpy, such that the array resembles a stack of 9x9 rank 2 arrays. Each of these arrays will be completely filled with ones, twos, threes, etc.

So, looking at one face of the cube we see ones, at the opposite face nines. And then at the sides columns where each column contains a number between 1 and 9.

Hopefully I've made this clear. I can do it for a rank 2 array, just not a rank 3 array.

This looks similar to a [previous question](http://stackoverflow.com/q/14513222/1191119) of yours — jorgeca, Jun 17 '13 at 21:19

score 3 · Accepted Answer · answered Jun 17 '13 at 21:11

If I understand correctly, you want:

c=np.ones((9,9,9))*np.arange(1,10)

To see this is what you want, you can slice for example c[:,:,0] to get a matrix of ones or c[:,:,-1] to get a matrix of nines. c[0,:,:] will be a matrix with rows increasing from 1 to 9.

Creating rank 3 numpy arrays

1 Answers1