MediumNLPnlp-186

Word Error Rate

Problem

Return the word error rate: the Levenshtein distance between reference and hypothesis token lists divided by the reference length.

An empty reference gives 0.0 for an empty hypothesis and 1.0 otherwise.

Hint

Use the standard edit-distance DP and normalize by the number of reference words.

case 1

in: [["a","b","c"],["a","b","c"]]

expected: 0

case 2

in: [["a","b","c"],["a","c"]]

expected: 0.3333333333333333

case 3

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

expected: 1

case 4

in: [[],[]]

expected: 0

case 5

in: [[],["x"]]

expected: 1

Starter codePython

def wer(reference, hypothesis):
    # Your code here
    pass

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