Python Sort Array

Question

Is it possible to sort an array with the smallest values on the bottom left and the highest on the top right?

For example :

I've this array :

[[-1.  1.  1.]
 [ 1.  1.  1.]
 [ 1.  1.  1.]
 [ 1.  1.  1.]
 [ 1.  1.  1.]
 [-1.  1.  1.]]

the sort would be like

[[1.  1.  1.]
 [ 1.  1.  1.]
 [ 1.  1.  1.]
 [ 1.  1.  1.]
 [-1.  1.  1.]
 [-1.  1.  1.]]

An other example :

[[-1.  1.  1.]
 [ 1.  1.  1.]
 [ 1.  1.  1.]
 [ 4.  3.  1.]
 [ 1.  1.  1.]
 [-1.  1.  1.]]

Would be :

[[ 1.  1.  4.]
 [ 1.  1.  3.]
 [ 1.  1.  1.]
 [ 1.  1.  1.]
 [-1.  1.  1.]
 [-1.  1.  1.]]

I Tried Numpy sort :

MyArray.sort()

But it seems not ordering this way.

You should make clear what type your "array" is. It does not seem to be a Python list. — Klaus D.
– Klaus D., Commented Mar 3, 2022 at 9:15
"But it seems not ordering this way." How does it order, and how is that different from what you want? Did you try reading the documentation? After reading the documentation, what steps did you take to refine the code, and how did it fail to improve the situation? If you could not find a way to solve the problem from reading the documentation, what exactly did you not understand about it? — Karl Knechtel
– Karl Knechtel, Commented Mar 3, 2022 at 9:15
@KlausD. Someone who writes an "array" in Numpy syntax and mentions trying to use Numpy sort, presumably has a Numpy array. I will add the numpy tag to the question. — Karl Knechtel
– Karl Knechtel, Commented Mar 3, 2022 at 9:16
"Smallest at bottom left and highest at top right" does not define the problem well enough. Does the sort propogate in a diagonal line from bottom left to top right? Or does it run up each column until it reaches the top right? — Andrew Guy
– Andrew Guy, Commented Mar 3, 2022 at 9:21

xjcl · Accepted Answer · 2022-03-03 09:36:18Z

import numpy as np

x = np.random.rand(6,3)

print(x)

def eiffel_tower_sort(a):
    b = a.flatten()
    b.sort()
    return np.flipud(b.reshape(a.shape, order='F'))

print(eiffel_tower_sort(x))

# randomized / unsorted
[[0.45884748 0.36774746 0.82728461]
 [0.46473908 0.22377053 0.43772489]
 [0.3596408  0.89647436 0.43567059]
 [0.10431368 0.06733271 0.26813345]
 [0.45886791 0.57112807 0.51780818]
 [0.27032551 0.77706324 0.32331996]]

# sorted
[[0.32331996 0.45886791 0.89647436]
 [0.27032551 0.45884748 0.82728461]
 [0.26813345 0.43772489 0.77706324]
 [0.22377053 0.43567059 0.57112807]
 [0.10431368 0.36774746 0.51780818]
 [0.06733271 0.3596408  0.46473908]]

Andrew Guy · Accepted Answer · 2022-03-03 09:49:05Z

This will get you a sorted array that is sorted from smallest to largest, running from the bottom left up each column till the top right.

import numpy as np

# Just some code to get a random, unsorted array
unsorted = np.arange(0,20)
np.random.shuffle(unsorted)
unsorted = unsorted.reshape((5,4))
print("Original Array:")
print(unsorted)

# Get the shape of the original array
array_shape = unsorted.shape
# Flatten, sort, reshape and rotate.
sorted_array = np.rot90(np.sort(unsorted.flatten()).reshape(array_shape[::-1]))

print("Sorted Array:")
print(sorted_array)

Output:

Original Array:
[[18 12  9 15]
 [ 1 19 17 13]
 [16  0 11  7]
 [ 2  6  5 14]
 [ 4  8  3 10]]
Sorted Array:
[[ 4  9 14 19]
 [ 3  8 13 18]
 [ 2  7 12 17]
 [ 1  6 11 16]
 [ 0  5 10 15]]

I'd have to give this some more thought to get a 'diagonally sorted' array.

Thank you, what you did works but I can't rotate this array (It will change the structure)
See my edits if you're referring to needing to keep the original array shape.

Collectives™ on Stack Overflow

Python Sort Array

2 Answers 2

Comments

2 Comments

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

2 Comments

Related