numpy array - choosing maximum based on multiple rows

Question

I want to write a function bestcolumn for a 2-dimensional numpy array that will do the following - choose the column with maximum value in row 0, if there are multiple values with the maximum in row 0, between them choose the column with maximum value in column 1 and then apply the same with column 2.

In the example below, 4 is the max value in row 0 but there are three columns 0,1,2 with the same value 4 so we search in row 1 cols 0,1,2. The max is 5 in col 1,2. We then search in col 2 for the tiebreaker and choose col 2

Is there a way to achieve this using numpy's vectorized operations (instead of using for loops and if/else statements)?

import numpy as np

A = np.asarray([[4,4,4,3],
                [3,5,5,7],
                [2,3,4,10]])

Not totally numpy but this would do: sorted([tuple(a) for a in A.T])[-1] gives (4,5,4). — Quang Hoang
– Quang Hoang, Commented May 24, 2020 at 14:23

V. Ayrat · Accepted Answer · 2020-05-24 15:06:25Z

1

One way is to use numpy.lexsort but it sort all columns not only find the maximal one.

np.lexsort(-A.T)[0]
# 2

answered May 24, 2020 at 15:06

V. Ayrat

2,74912 silver badges10 bronze badges

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

Collectives™ on Stack Overflow

numpy array - choosing maximum based on multiple rows

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