Multiple comparisons on an array of ints-CodePudding

I have an array of ints

mypos = np.array([10, 20, 30, 40, 50])

and a list of start & end positions, like this:

mydelims = [[5, 12],
            [15,31],
            [12,16],
            [22,69]]

I'd like to loop through mydelims and count how many positions are within each pair. Instinctively, I would write it like this:

for mypair in mydelims:
    print(sum(mypos>mypair[0] & mypos<mypair[1]))

However, python doesn't seem to handle this. I know mypos>42 will return an array of booleans, but what is the most efficient way to apply multiple comparisons to one array of ints?

CodePudding user response：

You were pretty close and looking for np.logical_and:

import numpy as np

mypos = np.array([10, 20, 30, 40, 50])

mydelims = [[5, 12],
            [15, 31],
            [12, 16],
            [22, 69]]

result = {
    tuple(mypair): sum(np.logical_and(mypos > mypair[0], mypos < mypair[1])) 
    for mypair in mydelims
}
print(result)

Output:

{(5, 12): 1, (15, 31): 2, (12, 16): 0, (22, 69): 3}

Note that you're treating the delims as non-inclusive - which I assumed was intentional. I.e. neither 5 nor 12 fall within [5, 12].