MediumNLPnlp-388

BLEU Clipped Unigram Precision

Problem

Clipped unigram precision counts each candidate token at most as many times as it appears in the reference, then divides by the candidate length.

Given the candidate tokens and reference tokens, return the precision.

Hint

Clip each candidate token count against the reference count.

case 1

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

expected: 0.6666666666666666

case 2

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

expected: 0

case 3

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

expected: 1

Starter codePython

def bleu_clipped_unigram_precision(candidate, reference):
    # Your code here
    pass

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