Transformer+YOLOv7排雷记录

来自于USTC的2022秋季数字图像处理project

本课程设计基于cuda 11.7和YOLO v7,引入了Swin transformer V2和BOTNET,COTNET,做了比较~

一,RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!

modify:在报错的地方定义device

device = torch.device('cuda:0')

issue:目前Swin transformer V2还不能这样修改,我是在CPU上跑得~

你可能感兴趣的:(transformer,深度学习,人工智能)