python - 后端 CPU 的预期对象，但参数 #2 得到后端 CUDA 'source'

Question

我尝试了其他答案，但错误没有被删除。与我收到的其他问题的区别在于，错误中使用的最后一个术语是“来源”，我在任何问题中都没有找到它。如果可能的话，还请解释错误的术语“来源”。在没有 CPU 的情况下运行代码也可以正常工作。

I am using Google Colab with GPU enabled.

import torch
from torch import nn
import syft as sy

hook = sy.TorchHook(torch)

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

model = nn.Sequential(nn.Linear(784,256),
                     nn.ReLU(),
                     nn.Linear(256,128),
                     nn.ReLU(),
                     nn.Linear(128,64),
                     nn.ReLU(),
                     nn.Linear(64,10),
                     nn.LogSoftmax(dim = 1))

model = model.to(device)

输出:

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-42-136ec343040a> in <module>()
      8                      nn.LogSoftmax(dim = 1))
      9 
---> 10 model = model.to(device)

3 frames
/usr/local/lib/python3.6/dist-packages/syft/frameworks/torch/hook/hook.py in data(self, new_data)
    368 
    369                 with torch.no_grad():
--> 370                     self.set_(new_data)
    371             return self
    372 

RuntimeError: Expected object of backend CPU but got backend CUDA for argument #2 'source'

Answer 1

此问题与 PySyft 有关。正如您在 Issue#1893 中看到的那样，当前workaround是要设置:

import torch
torch.set_default_tensor_type(torch.cuda.FloatTensor)

就在导入 torch 之后。

代码:

import torch
from torch import nn
torch.set_default_tensor_type(torch.cuda.FloatTensor)  # <-- workaround

import syft as sy
hook = sy.TorchHook(torch)

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

model = nn.Sequential(nn.Linear(784,256),
                     nn.ReLU(),
                     nn.Linear(256,128),
                     nn.ReLU(),
                     nn.Linear(128,64),
                     nn.ReLU(),
                     nn.Linear(64,10),
                     nn.LogSoftmax(dim = 1))

model = model.to(device)
print(model)

输出:

cuda
Sequential(
  (0): Linear(in_features=784, out_features=256, bias=True)
  (1): ReLU()
  (2): Linear(in_features=256, out_features=128, bias=True)
  (3): ReLU()
  (4): Linear(in_features=128, out_features=64, bias=True)
  (5): ReLU()
  (6): Linear(in_features=64, out_features=10, bias=True)
  (7): LogSoftmax()
)