python - Grouping 2D numpy array in average

Question

I am trying to group a numpy array into smaller size by taking average of the elements. Such as take average foreach 5x5 sub-arrays in a 100x100 array to create a 20x20 size array. As I have a huge data need to manipulate, is that an efficient way to do that?

Answer 1

I have tried this for smaller array, so test it with yours:

import numpy as np

nbig = 100
nsmall = 20
big = np.arange(nbig * nbig).reshape([nbig, nbig]) # 100x100

small = big.reshape([nsmall, nbig//nsmall, nsmall, nbig//nsmall]).mean(3).mean(1)

An example with 6x6 -> 3x3:

nbig = 6
nsmall = 3
big = np.arange(36).reshape([6,6])
array([[ 0,  1,  2,  3,  4,  5],
       [ 6,  7,  8,  9, 10, 11],
       [12, 13, 14, 15, 16, 17],
       [18, 19, 20, 21, 22, 23],
       [24, 25, 26, 27, 28, 29],
       [30, 31, 32, 33, 34, 35]])

small = big.reshape([nsmall, nbig//nsmall, nsmall, nbig//nsmall]).mean(3).mean(1)

array([[  3.5,   5.5,   7.5],
       [ 15.5,  17.5,  19.5],
       [ 27.5,  29.5,  31.5]])