
        
        
            
                
                
                    import numpy as np
x = np.array(['A','B','C','D'])
y = np.array([4,3,2,1])

Ascending order

To sort array y in ascending order and also sort array x maintaining the pair relationships between elements in the two arrays:


        
        
            
                
                
                    sort = np.argsort(y)
print(y[sort])
print(x[sort])
                
            
            [1 2 3 4]
['D' 'C' 'B' 'A']

First we get the integer indices that would result in array y being sorted in ascending order, and store them to variable sort. If we actually printed this variable we would see an array [3 2 1 0]. This means that to obtain an ascending sorted copy of array y, the element at index position 3 should come first, element at index position 2 should come second and so on.

We then take sort and apply it to both y and x. For both arrays we see the printed arrays have the element that was at index position 3 (1 and 'D' respectively) appear first, then have element that was at index position 2 (2 and 'C' respectively) appear next.

Descending order

To sort array x in descending order and also sort array y maintaining the pair relationships between elements in the two arrays:


        
        
            
                
                
                    sort = np.argsort(x)
print(x[~sort])
print(y[~sort])
                
            
            ['D' 'C' 'B' 'A']
[1 2 3 4]

Although np.argsort(~) does not natively support descending order, you can leverage negation (~) to implement the functionality. By negating the sort array here, the index position of the highest elements will now come first and lowest elements last.

Published by Arthur Yanagisawa

Edited by 0 others

Did you find this page useful?

thumb_up

thumb_down

Comment

Citation

Ask a question or leave a feedback...

thumb_up

thumb_down

chat_bubble_outline

settings

Enjoy our search

Hit / to insta-search docs and recipes!