edit unixy loss and evaluation

[amyburden]

idx = torch.argsort(position) position[idx] is sorted position from small to large from 0 to 1, idx_f = torch.arange(p).float().to(self.device) is 0 to p uni_l2 = torch.mean(torch.pow(position[idx] - (idx_f / p), 2)) is describe as in paper. but weight of uni_l2 may need changes.

evaluation part: point on warp image should be filter when they do not exist on original image