【已解决】pytorch cnn-text-classification 代码运行报错

在运行 Shawn1993/cnn-text-classification-pytorch的代码时，出现了错误。

Environment:
python:3.5
pytorch: 1.0.0.dev20190122

a.问题：

Traceback (most recent call last):
  File "/cnn-text-classification-pytorch/main.py", line 112, in <module>
    train.train(train_iter, dev_iter, cnn, args)
  File "/cnn-text-classification-pytorch/train.py", line 25, in train
    feature.data.t_(), target.data.sub_(1)  # batch first, index align
RuntimeError: set_storage_offset is not allowed on Tensor created from .data or .detach()

Process finished with exit code 1

a.解决方法：

feature.data.t_() ： 原地进行转置
target.data.sub_(1) ：把target中的每个值都减1

替换为：

 feature = feature.data.t()   # x.t() x是不变的，所以重新赋值
 target = target.data.sub(1)  # x.sub() x是不变的，所以重新赋值

b.问题

Traceback (most recent call last):
  File "/cnn-text-classification-pytorch/main.py", line 112, in <module>
    train.train(train_iter, dev_iter, cnn, args)
  File "/cnn-text-classification-pytorch/train.py", line 43, in train
    loss.data[0],
IndexError: invalid index of a 0-dim tensor. Use tensor.item() to convert a 0-dim tensor to a Python number

Process finished with exit code 1

b.解决方法

loss.data[0]

替换为：

loss.item()

参考资料

UserWarning: invalid index of a 0-dim tensor
pytorch tensor 介绍

x.sub_()
x.t_()