Target Encoding with Smoothing

Problem

Target-encode categories with additive smoothing: encoding(c) = (sum of targets in c + m * prior) / (count_c + m), where prior is the global target mean and m is the smoothing strength.

Return a dict mapping str(category) to encoding. Empty input returns {}.

Hint

Smoothing pulls rare categories toward the global mean; keys are stringified labels.

Test cases

case 1

in: [["a","a","b","b","b"],[1,0,1,1,0],2]

expected: {"a":0.55,"b":0.64}

case 2

in: [["a","a","b","b","b"],[1,0,1,1,0],0]

expected: {"a":0.5,"b":0.6666666666666666}

case 3

in: [["x"],[2],1]

expected: {"x":2}

case 4

in: [[],[],1]

expected: {}

Starter codePython

def target_encode(categories, targets, smoothing):
    # Your code here
    pass

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