Tombs
Tombs
channels
> 您好，我对您的模型很感兴趣，在使用时有个疑问，MHN的输出是return [stage2_img + x3_img, x_s1]，使用这个输出和使用单个的输出相比，是会有一些优势吗 没有，只是为了做消融实验，对两个输出做loss
> when I was training your code, I found the loss value became the nan. And the PSNR value also became the nan. Moreover, the best model didn't change after...
> 你好 训练时 损失一直是负的 梯度是零 请问怎么解决 https://github.com/Tombs98/MHNet/issues/6#issuecomment-1643864954
> Hello, > > I hope you’re doing well. I have a question regarding the training process described in your article. I followed the instructions and started training the model,...
> Your article has been especially helpful to me，However, it seems that CharbonnierLoss and EdgeLoss that you defined in loss.py are not reflected in train.py. We look forward to hearing...
> Your work has been incredibly helpful to me, and I am very grateful. I was wondering if you could provide some information on the expected training time and required...
> 请问会发布test_res.py 这个文件吗 Details of our training will be released in the future
> You can update the channel of 64 in the network. As far as I know, the larger the patch size, the better the final effect will be.(you also can...
> Thanks for your reply. Is it channel? Otherwise width? Do u mean this parameter?  What do u think about the total training epochs? Details of our training will...