Reliability Diagram Binning

Problem

Return the bin index for each probability using int(p * num_bins), clipped to [0, num_bins - 1] so that 1.0 falls in the final bin. Return an empty list when num_bins <= 0.

Test cases

case 1

in: [[0,0.5,0.999,1],5]

expected: [0,2,4,4]

case 2

in: [[],4]

expected: []

case 3

in: [[0.2],0]

expected: []

case 4

in: [[0.4,0.6],2]

expected: [0,1]

Starter codePython

def reliability_bins(probs, num_bins):
    # Your code here
    pass

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