解决方案RuntimeError: save_for_backward can only save input or output tensors, but argument 0

版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/github_39611196/article/details/82998092

本篇博客主要解决Torch在运行过程中出现的如下问题：

Traceback (most recent call last):
  File "D:/Python27/Lib/site-packages/django/bin/pytorchlearn/name_class.py", line 133, in <module>
    output, next_hidden = rnn(input[0], hidden)
  File "D:\Users\Seavan_CC\Anaconda3\lib\site-packages\torch\nn\modules\module.py", line 206, in __call__
    result = self.forward(*input, **kwargs)
  File "D:/Python27/Lib/site-packages/django/bin/pytorchlearn/name_class.py", line 106, in forward
    hidden = self.i2h(combined)  # 计算隐藏层参数
  File "D:\Users\Seavan_CC\Anaconda3\lib\site-packages\torch\nn\modules\module.py", line 206, in __call__
    result = self.forward(*input, **kwargs)
  File "D:\Users\Seavan_CC\Anaconda3\lib\site-packages\torch\nn\modules\linear.py", line 54, in forward
    return self._backend.Linear.apply(input, self.weight, self.bias)
RuntimeError: save_for_backward can only save input or output tensors, but argument 0 doesn't satisfy this condition

解决方案：

导入torch.autograd.Variable,并使用Variable将输入变量转换为input tensor，具体代码如下所示：

将

output, next_hidden = rnn(input, hidden)

转换为

from torch.autograd import Variable
output, next_hidden = rnn(Variable(input, Variable(hidden))