EasyNLPnlp-303

METEOR Weighted F-mean

Problem

Return the METEOR weighted F-mean of precision and recall:

Fmean = (1 + beta ** 2) * precision * recall / (recall + beta ** 2 * precision)

beta controls the recall weight, with METEOR traditionally using beta = 3. Return 0.0 when the denominator is 0.

Hint

When beta is 0 the formula reduces to precision.

case 1

in: [0.5,0.5,3]

expected: 0.5

case 2

in: [1,0.5,3]

expected: 0.5263157894736842

case 3

in: [0,0.5,1]

expected: 0

case 4

in: [0.5,0.5,0]

expected: 0.5

case 5

in: [1,1,2]

expected: 1

Starter codePython

def meteor_weighted_fmean(precision, recall, beta):
    # Your code here
    pass

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