Need help for DRCN model

huangzehao / caffe-vdsr

A Caffe-based implementation of very deep convolution network for image super-resolution

MIT License

273 stars 134 forks source link

Need help for DRCN model #13

Closed BaiYu0120 closed 8 years ago

BaiYu0120 commented 8 years ago

Hi, my inplement of DRCN model did not show better performance than VDSR, can you help me to check the net.prototxt? baiyupc@163.com,this is my email address , can you send me your email address?

BaiYu0120 commented 8 years ago

Email has been sent.

huangzehao commented 8 years ago

Hi, maybe there are some problems in the loss setting. Check this http://caffe.berkeleyvision.org/tutorial/loss.html and set specific loss_weight in each loss layer. layer { name: "loss" type: "SoftmaxWithLoss" bottom: "pred" bottom: "label" top: "loss" loss_weight: 1 }