res-loglikelihood-regression
res-loglikelihood-regression copied to clipboard
Why fc_coord is Linear() Instead of nn.linear in Regression_nf_3d.py?
- Why design a separate linear layer for the output of the coordinate mean, and what is the difference between it and the ordinary fully connected layer?
- Why is this happening? x_norm = torch.norm(x, dim=1, keepdim=True) y = y / x_norm