LSTM PyTorch中带有LongTensor的RuntimeError

Question

我正在尝试使seq2seq自动编码器，并且我想使用预训练的嵌入，所以我不在自动编码器体系结构中使用它。 LSTM出现了奇怪的错误：

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-91-fd376d45585e> in <module>
     18     for ei in range(input_length):
     19         encoder_output, encoder_hidden, encoder_cell = encoder(
---> 20             input_tensor[ei], encoder_hidden, encoder_cell)
     21         encoder_outputs[ei] = encoder_output[0, 0]
     22 

~/miniconda3/envs/simple_code/lib/python3.7/site-packages/torch/nn/modules/module.py in __call__(self, *input, **kwargs)
    545             result = self._slow_forward(*input, **kwargs)
    546         else:
--> 547             result = self.forward(*input, **kwargs)
    548         for hook in self._forward_hooks.values():
    549             hook_result = hook(self, input, result)

<ipython-input-89-012b4c3b2071> in forward(self, _input, hidden, cell)
      8         _input = _input.view(1,1,-1)
      9         print(type(_input), type(hidden), type(cell))
---> 10         output, (hidden, cell) = self.lstm(_input.view(1,1,-1).long(), (hidden.long(), cell.long()))
     11         return output, hidden, cell
     12 

~/miniconda3/envs/simple_code/lib/python3.7/site-packages/torch/nn/modules/module.py in __call__(self, *input, **kwargs)
    545             result = self._slow_forward(*input, **kwargs)
    546         else:
--> 547             result = self.forward(*input, **kwargs)
    548         for hook in self._forward_hooks.values():
    549             hook_result = hook(self, input, result)

~/miniconda3/envs/simple_code/lib/python3.7/site-packages/torch/nn/modules/rnn.py in forward(self, input, hx)
    562             return self.forward_packed(input, hx)
    563         else:
--> 564             return self.forward_tensor(input, hx)
    565 
    566 class GRU(RNNBase):

~/miniconda3/envs/simple_code/lib/python3.7/site-packages/torch/nn/modules/rnn.py in forward_tensor(self, input, hx)
    541         unsorted_indices = None
    542 
--> 543         output, hidden = self.forward_impl(input, hx, batch_sizes, max_batch_size, sorted_indices)
    544 
    545         return output, self.permute_hidden(hidden, unsorted_indices)

~/miniconda3/envs/simple_code/lib/python3.7/site-packages/torch/nn/modules/rnn.py in forward_impl(self, input, hx, batch_sizes, max_batch_size, sorted_indices)
    524         if batch_sizes is None:
    525             result = _VF.lstm(input, hx, self._get_flat_weights(), self.bias, self.num_layers,
--> 526                               self.dropout, self.training, self.bidirectional, self.batch_first)
    527         else:
    528             result = _VF.lstm(input, batch_sizes, hx, self._get_flat_weights(), self.bias,

RuntimeError: Expected object of scalar type Long but got scalar type Float for argument #2 'mat2'

有一个字符串，出现错误：

output, (hidden, cell) = self.lstm(_input.view(1,1,-1).long(), (hidden.long(), cell.long()))

我不明白，为什么会发生错误，因为我将所有张量都强制转换为Long类型