Mixup Label Blend

Problem

Interpolate two label distributions with the mixup lambda.

Return lam * y_a + (1 - lam) * y_b elementwise, rounded to 4 decimal places, where lam is the mixing weight applied to the first label vector.

case 1

in: [[1,0],[0,1],0.7]

expected: [0.7,0.3]

case 2

in: [[0.2,0.8],[0.9,0.1],0.25]

expected: [0.725,0.275]

case 3

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

expected: [1,0,0]

Starter codePython

def mixup_labels(y_a, y_b, lam):
    # Your code here
    pass

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