Yuandong Pu

icon indicating a website link https://www.sjtu.edu.cn

icon company Shanghai Jiao Tong University, @Xpixel and @OpenGVLab icon location Shanghai icon location A member of @OpenGVLab, Shanghai AI Laboratory

Results 3 comments of Yuandong Pu

About the training?

These functions are not well imported. You can try this. It solves my problem. from torch.utils import data as data from torchvision.transforms.functional import normalize from basicsr.data.data_util import (paired_paths_from_folder, paired_DP_paths_from_folder, paired_paths_from_lmdb,...

About the training?

> I have tried your method.But there is a new question. Do you have a solution to this? Thanks~ > > File "Restormer/basicsr/data/paired_image_dataset.py", line 223, in **getitem** gt_path = self.paths[index]['gt_path']...

Shape invalid for input of size

Hello, I've come across the same error. I discovered that this runtime issue stems from a shape mismatch between 'q' and 'k' in the cross-attention module. The batch sizes differ...