RuntimeError：只有浮点 dtype 的张量才能需要渐变

Question

RuntimeError: 只有浮点 dtype 的张量可以需要渐变
从

收到以下错误

输入 = 变量（preprocessed_img, requires_grad = True）

img=train_loader.dataset.data[0]
print(type(img))
img_tensor = torch.tensor(img)
preprocess_image(img)

> def preprocess_image(img):    means=[0.485, 0.456, 0.406]     stds=[0.229,
> 0.224, 0.225]
> 
>   preprocessed_img = img.copy()[: , :, ::-1]  for i in range(3):
>       preprocessed_img[:, :, i] = preprocessed_img[:, :, i] - means[i]
>       preprocessed_img[:, :, i] = preprocessed_img[:, :, i] / stds[i]
>   preprocessed_img = \
>       np.ascontiguousarray(np.transpose(preprocessed_img, (2, 0, 1)))
>   preprocessed_img = torch.from_numpy(preprocessed_img)
>   preprocessed_img.unsqueeze_(0)  input = Variable(preprocessed_img,
> requires_grad = True)     return input

Answer 1

使用torch.Tensor(img) 代替torch.tensor(img)。这应该可以解决您的问题。