content-aware-gan-compression Question about the defintion of Pruning loss

Question about the defintion of Pruning loss

Open tanbuzheng opened this issue 2 years ago • 1 comments

loss = torch.sum(torch.abs(noisy_img - img_tensor)) I hope to know why define pruning loss like above and the reason for minimizing this loss.

Dec 28 '22 14:12 tanbuzheng

First, it's a back-propagate-able loss. Second, we use it to evaluate channels' sensitivity for the content region noises. Please refer to the paper Sec. 3.1.2 for more details.

Dec 28 '22 22:12 lychenyoko

content-aware-gan-compression content-aware-gan-compression copied to clipboard

Question about the defintion of Pruning loss

content-aware-gan-compression
content-aware-gan-compression copied to clipboard