RuntimeError: Expected object of scalar type Double but got scalar type Float for argument #2 'mat2'

问题：

RuntimeError: Expected object of scalar type Double but got scalar type Float for argument #2 ‘mat2’

异常代码行：

prediction = net.forward(b_x)

解决方法：

在代码前添加一行，将输入的数据转成是dtype=torch.float32的。

b_x = torch.tensor(b_x, dtype=torch.float32)
prediction = net.forward(b_x)

参考链接

https://github.com/pytorch/pytorch/issues/2138、

https://pytorch.org/docs/stable/tensors.html?highlight=float#torch.Tensor.float

# if you use the DataLoader you can avoid any miscommunication regarding built-in data structures for ATen

X = torch.from_numpy(X).double()
y = torch.from_numpy(y).double()

train_data = torch.utils.data.TensorDataset(X, y)
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, shuffle=True)

X = Variable(X).float()
y = Variable(y).type(torch.LongTensor)

有用请点赞，谢谢

完