Transform 3D numpy array in boolean 2D numpy array with nonzero condition

Question

Imagine we have the following 3D array:

[[[0, 0, 0],
 [0, 0, 0],
 [0, 0, 0]],

 [[1, 2, 1],
 [0, 0, 0],
 [4, 2, 1]],

 [[0, 0, 0],
 [1, 1, 3],
 [0, 0, 0]]]

I want to reduce the dimension of the array and obtain True or False whenever each element in axis=2 is different or equal to [0, 0, 0], respectively.

Desired 2D output for the previous example array:

[[False, False, False]
 [True, False, True],
 [False, True, False]]

We pass from 3x3x3 int/float array to a 3x3 boolean array.

Andrej Kesely · Accepted Answer · 2022-07-26 20:59:19Z

2

Try:

print(arr[:, :, 1] != 0)  # 1 means second column, change to 0 for first, 2 for third column

Prints:

[[False False False]
 [ True False  True]
 [False  True False]]

Note: I used basing numpy indexing of ndarray. [:, :, 1] means that I want values from every 2D matrix, every row and second (1) column. Then compare these values to 0 to obtain boolean matrix.

edited Jul 26, 2022 at 20:59

answered Jul 26, 2022 at 20:30

Andrej Kesely

196k15 gold badges60 silver badges105 bronze badges

Sign up to request clarification or add additional context in comments.

10 Comments

Janikas Over a year ago

It did not work :(

Andrej Kesely Over a year ago

@Janikas Can you explain more? What did not work?

Janikas Over a year ago

Sorry, my bad! It worked! :)

Dani Mesejo Over a year ago

It would be nice if you explain how this works. To the best of my knowledge arr[:, :, 1] just takes the element at position 1 for each array at axis 2

Andrej Kesely Over a year ago

@DaniMesejo Yes, I added short explanation.

|

mozway · Accepted Answer · 2022-07-26 21:14:49Z

2

IIUC you want to compare the last dimension to [0,0,0], returning False if all values are different, True otherwise.

Assuming a the 3D array:

out = (a != [0,0,0]).any(2)

Alternative:

out = ~(a == [0,0,0]).all(2)

Output:

array([[False, False, False],
       [ True, False,  True],
       [False,  True, False]])

answered Jul 26, 2022 at 21:14

mozway

267k13 gold badges56 silver badges106 bronze badges

2 Comments

Onyambu Over a year ago

(test != [0,0,0]).all(2) should also work. I believe the notion of each element different needs to be addressed by all rather than any

mozway Over a year ago

@onyambu yes could be, the example is unfortunately ambiguous (all zero or all non-zero)

Tzn · Accepted Answer · 2022-07-27 04:34:01Z

1

import numpy as np

arr = np.array([
    [
        [0, 0, 0],
        [0, 0, 0],
        [0, 0, 0],
    ],

    [
        [1, 2, 1],
        [0, 0, 0],
        [4, 2, 1],
    ],

    [
        [0, 0, 0],
        [1, 1, 3],
        [0, 0, 0],
    ]
])

answer = (arr != 0).any(axis=2)

print(answer)
# [[False False False]
#  [ True False  True]
#  [False  True False]]

answered Jul 27, 2022 at 4:34

Tzn

7826 silver badges17 bronze badges

Comments

Ricardo · Accepted Answer · 2022-07-26 20:38:11Z

0

 >>> test = [[[0, 0, 0],
 [0, 0, 0],
 [0, 0, 0]],

 [[1, 2, 1],
 [0, 0, 0],
 [4, 2, 1]],

 [[0, 0, 0],
 [1, 1, 3],
 [0, 0, 0]]]
 >>> [[not ee == [0, 0, 0] for ee in e] for e in test]

answered Jul 26, 2022 at 20:38

Ricardo

6914 silver badges12 bronze badges

1 Comment

Ricardo Over a year ago

I used list instead of numpy array. You can convert the numpy array to list first by "test.tolist()"

Collectives™ on Stack Overflow

Transform 3D numpy array in boolean 2D numpy array with nonzero condition

4 Answers 4

10 Comments

2 Comments

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

10 Comments

2 Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related