How to sum a numpy along the row axis by including only certain values per row according to variable length indices?

Question

For the following array:

array = [
  [1, 5, 6, 8, 10, 3],
  [3, 2, 4, 9, 11, 7],
  [8, 0, 9, 6, 23, 4]
]

How could we sum the elements (per row) as indicated by these indices:

indices = [
  [2, 4, 5],
  [1, 3],
  [4]
]

that is to say that:

for the first row only the values on indices [2, 4, 5] will be considered when summing up -> (6 + 10 + 3)
for the second row only the values on indices [1, 3] will be considered when summing up -> (2 + 9)
and so on

Output:

array([19, 11, 23])

The output has the same shape as if we did array.sum(axis=1) but not every value is included. Instead, the participants of each row are determined by the indices array.

I have thought of creating a mask for that purpose, but I did not know how to pass the indices to it.

How about a new array with 0 for the values you don't want to sum? — hpaulj
– hpaulj, Commented Nov 5, 2022 at 20:10

Mahdi F. · Accepted Answer · 2022-11-05 19:48:28Z

1

Try this:

arr = np.array(array)
out = np.array([arr[idx, ind].sum() for idx, ind in enumerate(indices)])
out

Output : array([19, 11, 23])

answered Nov 5, 2022 at 19:48

Mahdi F.

24.1k5 gold badges25 silver badges32 bronze badges

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

Collectives™ on Stack Overflow

How to sum a numpy along the row axis by including only certain values per row according to variable length indices?

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related