Receive Numpy mass indices

[Kadir]

Hello, everyone.

I have this Numpy mass:

array([16, 17, 1, 3, 8, 10, 2, 19, 14, 18, 15, 6, 11, 13, 5, 4, 9, 0, 7, 12])

I need to first get the indexes of each element together with the value, and then sort out the collected list of elements in such a way as to get something like that:

[19: 7, 18: 9, 17: 1, 16: 0, 15: 10 ...]

I mean, in the final list, I have to quantify by lowering the values from the first list, but I need to add the index on the base list for each element. And you need to do it the fastest way. I have an idea how to do it, but it's kind of hard to get out. Maybe numpy has a way faster?

[charlenef]

If you want to stay within the framework numpy and use its speed, it is possible to:

import numpy as np
arr = np.array([16, 17, 1, 3, 8, 10, 2, 19, 14, 18, 15, 6, 11, 13, 5, 4, 9, 0, 7, 12])

sorted_idx = arr.argsort()[::-1]

sorted_arr = arr[sorted_idx]

print(np.stack([sorted_arr, sorted_idx]))

Conclusion:

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

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

♪ numpy There's a sorting of mass indexes. This is where I get this sorting first, turn it over so that at the beginning there's a maximum, and then it's easy to get numbers from the index array. Well, if necessary, merge the indices and values into one set. Indexing in numpy It must be pretty fast.