Questions about training

Boyan-Lenin-Xu / GCResNet

PyTorch implementation of Graph Convolutional Networks in Feature Space for Image Deblurring and Super-resolution, IJCNN 2021.

12 stars 0 forks source link

Questions about training #3

Open STQ-AmadeusUser opened 2 years ago

STQ-AmadeusUser commented 2 years ago

Hello! I want to train your GCResNet for image deblurring. The options.xml that i utilize is default in github. I wonder how much the l1 loss will redeuce to if i use GoPro dataset after training 1000 epoches.