I have an array of indexes:

test_idxs = np.array([4, 2, 7, 5])

I also have an array of values (which is longer):

test_vals = np.array([13, 19, 31, 6, 21, 45, 98, 131, 11])

So I want to get an array with the length of the array of indexes, but with values from the array of values in the order of the array of indexes. In other words, I want to get something like this:

array([21, 31, 131, 45])

I know how to do this in a loop, but I don't know how to achieve this using Numpy tools.

I have an array of indexes:

test_idxs = np.array([4, 2, 7, 5])

I also have an array of values (which is longer):

test_vals = np.array([13, 19, 31, 6, 21, 45, 98, 131, 11])

So I want to get an array with the length of the array of indexes, but with values from the array of values in the order of the array of indexes. In other words, I want to get something like this:

array([21, 31, 131, 45])

I know how to do this in a loop, but I don't know how to achieve this using Numpy tools.

• python
• numpy

Share Improve this question Follow edited Nov 18, 2024 at 12:39 simon 5,65511 gold badge1616 silver badges2929 bronze badges asked Nov 18, 2024 at 12:11 IzaeDAIzaeDA 39766 bronze badges

Add a comment  | 

1 Answer 1

Sorted by: Reset to default 1

This is actually extremely simple with numpy, just index your test_vals array with test_idx (integer array indexing):

out = test_vals[test_idxs]

Output:

array([ 21,  31, 131,  45])

Note that this requires the indices to be valid. If you have indices that could be too high you would need to handle them explicitly.

Example:

test_idxs = np.array([4, 2, 9, 5])
test_vals = np.array([13, 19, 31, 6, 21, 45, 98, 131, 11])

out = np.where(test_idxs < len(test_vals),
               test_vals[np.clip(test_idxs, 0, len(test_vals)-1)],
               np.nan)

Output:

array([21., 31., nan, 45.])