CODEWITHSUNDEEP
pythonarraysnumpy
user3211991
user3211991

Reputation: 461

Basics of numpy where function, what does it do to the array?

I have seen the post Difference between nonzero(a), where(a) and argwhere(a). When to use which? and I don't really understand the use of the where function from numpy module.

For example I have this code

import numpy as np

Z =np.array( 
    [[1,0,1,1,0,0],
     [0,0,0,1,0,0],
     [0,1,0,1,0,0],
     [0,0,1,1,0,0],
     [0,1,0,0,0,0],
     [0,0,0,0,0,0]])
print Z
print np.where(Z)

Which gives:

(array([0, 0, 0, 1, 2, 2, 3, 3, 4], dtype=int64), 
 array([0, 2, 3, 3, 1, 3, 2, 3, 1], dtype=int64))

The definition of where function is: Return elements, either from x or y, depending on condition. But it doesn't also makes sense to me

So what does the output exactly mean?

Upvotes: 3

Views: 3667

Answers (1)

Matt
Matt

Reputation: 17649

np.where returns indices where a given condition is met. In your case, you're asking for the indices where the value in Z is not 0 (e.g. Python considers any non-0 value as True). Which for Z results in:

(0, 0) # top left
(0, 2) # third element in the first row
(0, 3) # fourth element in the first row
(1, 3) # fourth element in the second row
...    # and so on

np.where starts to make sense in the following scenarios:

a = np.arange(10)
np.where(a > 5) # give me all indices where the value of a is bigger than 5
# a > 5 is a boolean mask like [False, False, ..., True, True, True]
# (array([6, 7, 8, 9], dtype=int64),)

Hope that helps.

Upvotes: 3

Related Questions