Macro F1 Score

Problem

Compute macro-averaged F1 across the classes present in y_true.

For each class compute precision, recall, and F1 = 2*p*r/(p+r), using 0.0 whenever a denominator is zero. Return the unweighted mean of the per-class F1 values. Empty input returns 0.0.

Hint

Treat each class as positive versus the rest, then average the F1 scores equally.

Test cases

case 1

in: [[0,1,2,0,1,2],[0,1,2,0,1,2]]

expected: 1

case 2

in: [[0,0,1,1,2,2],[0,1,1,2,2,0]]

expected: 0.5

case 3

in: [[0,1,0,1],[1,0,1,0]]

expected: 0

case 4

in: [[0,0,0],[0,0,1]]

expected: 0.8

Starter codePython

def macro_f1(y_true, y_pred):
    # Your code here
    pass

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