Get values in numpy array either side of a specified value

Question

I have a numpy array, provided at random, which for this example looks like:

a = [10, 8, 6, 4, 2, -2, -4, -6, -8, -10, 1]

ideally, in this example, the values would be between -10 and 10 but this cannot be guaranteed (as above).

I want to retrive the 2 values closest to zero, such that:

b = a[a > 0][-1] 

c = a[a < 0][0]

which would ideally return me the values of 2 and -2. However, the 1 value is included in the slice in b and i get returned the values of 1 and -2.

Is there a way in numpy to retrieve the values immediately 'next' to zero?

Its worth noting that whilst I always want to split the array at 0, the array could be any length and I could have an uneven number of positive and negative values in the array (i.e. [5, 4, 3, 2, 1, 0, -1])

A real world example is:

I want the yellow and green position but get returned the blue and green position instead, as the data crosses back over zero from -ve to +ve

Thought I could use the index position of 0 and go from there, then realized I didn't have a 0 value. Ive edited the post to reflect this (thanks to those that also suggested this) — GeoMonkey
– GeoMonkey, Commented Jun 28, 2016 at 0:37
Not clear about the definition. Firstly, shouldn't 1 be the closest value that is close to 0 in the first example; in the second example should you return 0 or 1 -1?; — akuiper
– akuiper, Commented Jun 28, 2016 at 0:45
What should be returned in your first example, [2, -2], [-10, 1], or both? Can we assume that the numbers are monotonically decreasing? — Tonechas
– Tonechas, Commented Jun 28, 2016 at 0:51
I need the values of -2 and 2 to be returned where the value of 1 is an anomaly (in the 2nd example I would want 1 and -1). It can't be assumed that the values are monotonically decreasing. Will add a figure — GeoMonkey
– GeoMonkey, Commented Jun 28, 2016 at 1:02

Tonechas · Accepted Answer · 2016-06-28 01:26:35Z

2

This function should do the job:

import numpy as np
def my_func(x):
    left = np.where(x[:-1]>0)[0][-1]
    right = 1 + np.where(x[1:]<0)[0][0]
    return x[left], x[right]

Demo:

>>> a = np.array([10, 8, 6, 4, 2, -2, -4, -6, -8, -10, 1])
>>> b = np.array([5, 4, 3, 2, 1, 0, -1])
>>> my_func(a)
(2, -2)
>>> my_func(b)
(1, -1)

edited Jun 28, 2016 at 1:26

answered Jun 28, 2016 at 0:30

Tonechas

13.8k16 gold badges52 silver badges85 bronze badges

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

1 Comment

GeoMonkey Over a year ago

It works. Hopefully there are no other scenarios that creep up and break it. I will test it and see

Collectives™ on Stack Overflow

Get values in numpy array either side of a specified value

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related