Perceptual Loss Across Layers

Problem

Compute a weighted perceptual loss across several feature layers.

For each layer take the mean squared error between the two feature maps and weight it by weights[layer]. Return the sum of the weighted layer losses rounded to 4 decimal places.

Test cases

case 1

in: [[[[1,2]]],[[[1,4]]],[1]]

expected: 2

case 2

in: [[[[1,2]]],[[[1,4]]],[0.5]]

expected: 1

case 3

in: [[[[1,1]],[[2,2]]],[[[1,2]],[[2,4]]],[1,2]]

expected: 4.5

case 4

in: [[[[0]]],[[[0]]],[3]]

expected: 0

Starter codePython

def perceptual_loss(features_a, features_b, weights):
    # Your code here
    pass

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