How to effeciently perform multiple element swaps with a numpy array?

Question

Within a numpy array of 48 elements, I have a series of swaps that I need to make.

34 -> 21 -> 42 -> 26 -> 34
36 -> 19 -> 44 -> 28 -> 36
39 -> 16 -> 47 -> 31 -> 39

where x -> y means the element at index x must go to the yth index. I'm trying to think of an efficient way to do this since the numbers are quite random and this is just one of the series of swaps that I need to make.

For example: Given the following array

original = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']

I want to make the following swaps

1 -> 4 -> 6 -> 1

so I end up with

swapped = ['a', 'g', 'c', 'd', 'b', 'f', 'e', 'h']

so the 1st index element went to the 4th index, the 4th to the 6th, and the 6th to the 1st.

What format will your swap indices be in? Will they be hardcoded or dynamic? — ThisIsAQuestion
– ThisIsAQuestion, Commented Nov 4, 2020 at 16:44

Homan · Accepted Answer · 2020-11-04 16:43:56Z

2

This is probably the most efficient way to perform a single of those sequences:

swaps = [34, 21, 42, 26, 34]
arr = np.arange(48)

from_idx = lst[:-1]
to_idx = lst[1:]
arr[to_idx] = arr[from_idx]

answered Nov 4, 2020 at 16:43

Homan

1656 bronze badges

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

Comments

Al W · Accepted Answer · 2020-11-04 16:38:34Z

0

Perhaps something like this.

def swap(input, index):
    start_index = index[0]
    for next_index in range(1, len(index)):
       input[start_index], input[next_index] = input[next_index] , input[start_index]
    
    return(input)

if __name__ == '__main__':
    input = ['A', 'B', 'C', 'D', 'E', 'F', 'G']
    index = [2,3,6,2]

    output = swap(input, index)
    print(output)

edited Nov 4, 2020 at 16:38

answered Nov 4, 2020 at 16:31

Al W

7,7611 gold badge21 silver badges41 bronze badges

Collectives™ on Stack Overflow

How to effeciently perform multiple element swaps with a numpy array?

2 Answers 2

Comments

Comments

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Related