Cohen's Kappa

Problem

Compute Cohen's kappa: (p_o - p_e) / (1 - p_e), where p_o is the observed agreement and p_e = sum over the union of labels of (true_count_c / n) * (pred_count_c / n).

If p_e == 1.0 or the input is empty, return 0.0.

Hint

Kappa measures agreement above the agreement expected by chance.

Test cases

case 1

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

expected: 1

case 2

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

expected: 0

case 3

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

expected: -0.3333333333333333

case 4

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

expected: 0

Starter codePython

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

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