Denormalize

Problem

Invert per-channel normalization.

Return v * std[c] + mean[c] for every channel value, rounded to 4 decimal places. No clamping is applied.

case 1

in: [[[[-2,0,2]]],[128,128,128],[64,64,64]]

expected: [[[0,128,256]]]

case 2

in: [[[[0,0],[4,2]]],[10,20],[5,10]]

expected: [[[10,20],[30,40]]]

case 3

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

expected: [[[1,0]]]

Starter codePython

def denormalize(image, mean, std):
    # Your code here
    pass

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