count the number of occurance of each one hot code

Question

I have a list of numpy arrays (one-hot represantation) like the example bellow, I want to count the number of occurances of each one-hot code.

[0 0 1 0 0 0 0 0 0 0]
[0 0 1 0 0 0 0 0 0 0]
[0 1 0 0 0 0 0 0 0 0]
[0 0 0 0 0 1 0 0 0 0]
[0 1 0 0 0 0 0 0 0 0]
[0 0 0 0 1 0 0 0 0 0]
[0 0 0 0 0 0 0 0 0 1]
[0 0 0 0 1 0 0 0 0 0]
[1 0 0 0 0 0 0 0 0 0]
[0 0 0 1 0 0 0 0 0 0]
[0 1 0 0 0 0 0 0 0 0]
[0 0 0 0 0 0 0 0 0 1]

Edit : Expected output :

[1 0 0 0 0 0 0 0 0 0] ==> 1 occurrence
[0 0 1 0 0 0 0 0 0 0] ==> 2 occurrences
[0 1 0 0 0 0 0 0 0 0] ==> 3 occurrences
[0 0 0 0 0 1 0 0 0 0] ==> 1 occurrence
[0 0 0 0 1 0 0 0 0 0] ==> 2 occurrences
[0 0 0 0 0 0 0 0 0 1] ==> 2 occurrences

Answer 1

I think you can get the result you seek:

[1 3 2 1 2 1 0 0 0 2]

indicating the count of occurrences of one hot in that position via a simple column-wise sum using ndarray.sum():

import numpy
data = numpy.array([
    [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
    [0, 0, 1, 0, 0, 0, 0, 0, 0, 0],
    [0, 1, 0, 0, 0, 0, 0, 0, 0, 0],
    [0, 0, 0, 0, 0, 1, 0, 0, 0, 0],
    [0, 1, 0, 0, 0, 0, 0, 0, 0, 0],
    [0, 0, 0, 0, 1, 0, 0, 0, 0, 0],
    [0, 0, 0, 0, 0, 0, 0, 0, 0, 1],
    [0, 0, 0, 0, 1, 0, 0, 0, 0, 0],
    [1, 0, 0, 0, 0, 0, 0, 0, 0, 0],
    [0, 0, 0, 1, 0, 0, 0, 0, 0, 0],
    [0, 1, 0, 0, 0, 0, 0, 0, 0, 0],
    [0, 0, 0, 0, 0, 0, 0, 0, 0, 1],
])
print(numpy.ndarray.sum(data, axis=0))

or more compactly as just:

print(data.sum(axis=0))

both should give you:

[1 3 2 1 2 1 0 0 0 2]