Python 2D Numpy Array to 1D (sort of..) [duplicate]

Question

Simple question, I want to get a 1D numpy array.

Given a 2D array where each row contains a single '1' value, how can it be converted to a 1D array, consisting of the column index of the '1' in the 2D array

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

to

[2 2 1 1 1 1 0]

How would I do it in python? I don't know the terminology for it, do let me know the proper terms for such transformation.

Answer 1

You are looking for the index with the maximum value along the first axis:

>>> a.argmax(axis=1)
array([2, 2, 2, 1, 1, 1, 1, 0])

a.argmax(axis=None, out=None)

Return indices of the maximum values along the given axis.

If the other values are not necessarily less than one, filter for 1 first. This gives an array of True and False values. Now, use argmax():

>>> (a == 1).argmax(axis=1)
array([2, 2, 2, 1, 1, 1, 1, 0])

True acts like a 1 and False like a 0, because bool inherits from int.