How to add values to only certain index in Numpy 2D Matrix

Question

I have a 2d Matrix

matrix = np.array([[1,2],[3,4],[5,6]])
index = np.array([0, 1, 1])
add_value = np.array([1, 2, 3])

I want to add add_value to matrix but only to the elements corresponding to index in the index list. For example, 1 in add_value should be added to the first element in [1,2], which is 1, resulting in 2. So the output should be

np.array([[2,2],[3,6],[5,9]])

matrix[np.r_[:matrix.shape[0]], index] += add_value?

bb1
– bb1

2022-01-14 03:34:26 +00:00
Commented Jan 14, 2022 at 3:34 — bb1
– bb1, Commented Jan 14, 2022 at 3:34
you are a genius. have a lovely day

John Doyle
– John Doyle

2022-01-14 03:36:42 +00:00
Commented Jan 14, 2022 at 3:36 — John Doyle
– John Doyle, Commented Jan 14, 2022 at 3:36

mozway · Accepted Answer · 2022-01-14 04:02:48Z

2

Use a simple multi-dimensional indexing:

matrix[np.arange(matrix.shape[0]), index] += add_value

Or using python builtins:

matrix[tuple(zip(*enumerate(index)))] += add_value

Output:

array([[2, 2],
       [3, 6],
       [5, 9]])

answered Jan 14, 2022 at 4:02

mozway

267k13 gold badges56 silver badges106 bronze badges

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

Comments

Toms · Accepted Answer · 2022-01-14 03:56:41Z

0

for i, x in enumerate(add_value):
    matrix[i][index[i]] += x

edited Jan 14, 2022 at 3:56

answered Jan 14, 2022 at 3:43

Toms

2191 silver badge7 bronze badges

2 Comments

mozway Over a year ago

This will be slow on large arrays

Toms Over a year ago

Good to know thanks. For me it's the easiest way to understand what's happening.

Collectives™ on Stack Overflow

How to add values to only certain index in Numpy 2D Matrix

2 Answers 2

Comments

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related