How to change the value of an entire array based on one value in Python

Question

Let's say I have this array:

import numpy as np
np.array([[0,2,7],[-3,4,0],[12,10,12]])

I'm trying to change the values of one column if one or more values in the column are less than zero. The code should be general and not just for this array How do I set all of the values in the middle vector to -3?

Answer 1

You could use a mask to filter the columns where some values are negative. That same mask can then be used to also select the replacement value: in the example below the columns where at least one item is negative, will be replaced by columns where the values are all equal to that minimum value.

>>> a = np.array([[0,2,7],[-3,4,0],[12,10,12], [0, -1, 3]]).T
>>> a
array([[ 0, -3, 12,  0],
       [ 2,  4, 10, -1],
       [ 7,  0, 12,  3]])
>>> mask = np.any(a < 0, axis=0)
>>> a[:,mask] = np.min(a[:,mask], axis=0)
>>> a
array([[ 0, -3, 12, -1],
       [ 2, -3, 10, -1],
       [ 7, -3, 12, -1]])