Closed TongkunGuan closed 12 months ago
To compute proto_reg: contrast_norm = num_classes * np.log(num_classes) proto_sim = mean_feat.mm(mean.permute(1, 0).contiguous()) / contrast_temp loss = torch.sum(torch.softmax(proto_sim, dim=1).log()) / contrast_norm
reg_weight * proto_reg(feat, mean, contrast_temp, contrast_norm=contrast_norm)
Why loss need < 0? I don't understand it, Thanks!
To compute proto_reg: contrast_norm = num_classes * np.log(num_classes) proto_sim = mean_feat.mm(mean.permute(1, 0).contiguous()) / contrast_temp loss = torch.sum(torch.softmax(proto_sim, dim=1).log()) / contrast_norm
reg_weight * proto_reg(feat, mean, contrast_temp, contrast_norm=contrast_norm)
Why loss need < 0? I don't understand it, Thanks!