RuntimeError: Expected object of scalar type Long but got scalar type Float for argument #2 ‘target‘

l = np.random.beta(20, 20)
                index = torch.randperm(inputs.size(0))
                images_a, images_b = inputs, inputs[index]
                labels_a, labels_b = inputs, labels[index]
                mixed_images = l * images_a + (1 - l) * images_b
                mixed_images = mixed_images.to(device)
                labels_a = labels_a.to(device)
                labels_b = labels_b.to(device)
                optimizer.zero_grad()
                outputs = model(mixed_images)
                loss = l * loss_function(outputs, labels_a) + (1 - l) * loss_function(outputs, labels_b)

遇到问题不要慌~

报错在最后一行，意味着loss_function，参数类型有错误，所以可以把labels类型该为long类型

labels.long()，so Easy

But好像问题还没解决，仔细一看

labels_a, labels_b = inputs, labels[index]

这是什么鬼！

so：

labels_a, labels_b = labels, labels[index]；

好像不加long()也不报错了。。。

这类错误应该就是输入类型不对导致的，变换类型或者检查输入；

感谢收看，欢迎一键三连！😄