What is the meaning of the 2-th power of the variable 'Win' ?

[cjfcsjt]

Hi! Thanks for your great work! I'm a bit confused with the code in 'MGCN_FAUST.py' for k in range(31, -1, -(32//(K-1))): Win = torch.mm(torch.mm(torch.mm(V, torch.diag(clk[:, k])), torch.t(V)), torch.diag(torch.squeeze(A))) ** 2 Win = torch.nn.functional.normalize(Win, p=2, dim=0) ** 2 list.append(Win) Win = torch.stack(list, dim=0) It seems that it calculates the wavelet basis, but I do not follow the meaning of the 2-th power of the variable 'Win', Could you explain it please?

[yiqun-wang]

Hi, I just want to make the wavelet normalized, and the sum of the normalized wavelet is 1.