Calibration Intercept Lite

Problem

Return the calibration intercept estimate logit(observed_rate) - logit(mean_predicted) with logit(p) = ln(p/(1-p)). Return 0.0 when either input is outside (0, 1).

Hint

It compares the observed event rate with the mean predicted probability on the log-odds scale.

Test cases

case 1

in: [0.2,0.25]

expected: -0.2876820724517808

case 2

in: [0.1,0.1]

expected: 0

case 3

in: [0.5,0.2]

expected: 1.3862943611198906

case 4

in: [0,0.5]

expected: 0

case 5

in: [0.9,0.8]

expected: 0.8109302162163288

Starter codePython

import math
def calibration_intercept_lite(observed_rate, mean_predicted):
    # Your code here
    pass

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