RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the

RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same

出现这个问题是因为输入模型的tensor加载在cpu中，而模型却加载在cuda上。

解决方法：将输入tensor加载到cuda中，或者把模型加载到cpu

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

model = model.to(device)
img = img.to(device)

output = model(img)

或者：

model = model.cuda()
img = img.cuda()