I do not understand what this code is doing. X is a matrix which each row of it is an eigenvector of an image. What does the norm do?
# Take an eigenvector and make it into an image
def vecToImage(x, size = lfw_imageSize):
im = x/np.linalg.norm(x)
im = im*(256./np.max(im))
im.resize(*size)
return im