Pytorch学习笔记(6)——RNN隐藏状态不能参与梯度计算

报错内容：

1. CPU上运行报错：RuntimeError: Trying to backward through the graph a second time (or directly access saved variables after they have already been freed). Saved intermediate values of the graph are freed when you call .backward() or autograd.grad(). Specify retain_graph=True if you need to backward through the graph a second time or if you need to access saved variables after calling backward.
2. GPU上运行报错：RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!

报错原因： 直接采用 RNN 的隐藏状态参与到模型运算中。
解决方案： 在使用 RNN 的隐藏状态时，需要使用 .detach() 将隐藏状态从计算图中分离出来。