RNN

而RNN本质上就是linear layers。

即使RNN的输入数据是batch first，内部也会转为seq_len first。

def forward(self, input, hx=None):
    batch_sizes = None # is not packed, batch_sizes = None
    max_batch_size = input.size(0) if self.batch_first else input.size(1) # batch_size
    
    if hx is None: # 使用者可以不传输hidden, 自动创建全0的hidden
        num_directions = 2 if self.bidirectional else 1
        hx = torch.autograd.Variable(input.data.new(self.num_layers *
                                                    num_directions,
                                                    max_batch_size,
                                                    self.hidden_size).zero_())
        if self.mode == 'LSTM': # h_0, c_0
            hx = (hx, hx)
   
    flat_weight = None # if cpu

    func = self._backend.RNN( # self._backend = thnn_backend # backend = THNNFunctionBackend(), FunctionBackend
        self.mode,
        self.input_size,
        self.hidden_size,
        num_layers=self.num_layers,
        batch_first=self.batch_first,
        dropout=self.dropout,
        train=self.training,
        bidirectional=self.bidirectional,
        batch_sizes=batch_sizes,
        dropout_state=self.dropout_state,
        flat_weight=flat_weight
    )
    output, hidden = func(input, self.all_weights, hx)

    return output, hidden

　　可以看到，在训练RNN时，可以不传入  ，此时PyTorch会自动创建全0的  。

也可以对RNN的output添加一层全连接层实现与hidden的不同