oarriaga/face_classification

Understanding the code

j13mehul opened this issue · 1 comments

def preprocess_input(x, v2=True):
    x = x.astype('float32')
    x = x / 255.0
    if v2:
        x = x - 0.5
        x = x * 2.0
    return x

I know x/255 is normalizing the image.
What is the purpose of x-0.5 and x*2.0

Hi since the image is normalized [0, 1] you then move it to [-.5, .5] and the finally you scaled it by [-1 ,1].