报错

[yt-liang]

您好，LSTM-CNNS-CRF中，运行train.py，报错RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.IntTensor instead (while checking arguments for embedding)。显示是model（word，char，label）的问题

[ne7ermore]

确认下pytorch的版本是不是0.4.0

[GZC201314]

我的也是这个错误，使用的pytorch是0.4.0
RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.IntTensor instead (while checking arguments for embedding)。