LIME Kernel Weight

Problem

Compute the LIME locality kernel weight: exp(-distance^2 / width^2).

Return 0.0 when width <= 0.

Hint

The kernel decays with distance so nearby perturbations dominate the surrogate fit.

case 1

in: [2,2]

expected: 0.36787944117144233

case 2

in: [0,2]

expected: 1

case 3

in: [1,0]

expected: 0

case 4

in: [3,1]

expected: 0.00012340980408667956

Starter codePython

import math
def lime_kernel_weight(distance, width):
    # Your code here
    pass

4 test cases · run your code against them in the editor