Frank_Liu

line 455-459: for i in range(confusion.size()[0]): mu = confusion[i, :].sum().item() p = confusion[i, i].item() / mu if mu != 0. else 0 r = confusion[i, i].item() / mu if mu...