ValueError:Dimension must be equal ,but for rnn_layer/bidirection_rnn/fw/fw/while/fw/multi_rnn_cell/

具体的错误类型如下图所示

 

修改之前的代码是：
 

    def blstm_layer(self, embedding_chars):
        with tf.variable_scope('rnn_layer'):
            cell_fw, cell_bw = self._bi_dir_rnn()
            if self.num_layers > 1:
                cell_fw = rnn.MultiRNNCell([cell_fw] * self.num_layers, state_is_tuple=True)
                cell_bw = rnn.MultiRNNCell([cell_bw] * self.num_layers, state_is_tuple=True)
bedding_chars,
                                                         dtype=tf.float32)
            outputs = tf.concat(outputs, axis=2)
        return outputs

修改之后的代码是：

    def blstm_layer(self, embedding_chars):
        with tf.variable_scope('rnn_layer'):
            cell_fw, cell_bw = self._bi_dir_rnn()
            if self.num_layers > 1:
                # cell_fw = rnn.MultiRNNCell([cell_fw] * self.num_layers, state_is_tuple=True)
                # cell_bw = rnn.MultiRNNCell([cell_bw] * self.num_layers, state_is_tuple=True)
                cell_fw = rnn.MultiRNNCell([self._bi_dir_rnn()[0] for _ in range(self.num_layers)], state_is_tuple=True)
                cell_bw = rnn.MultiRNNCell([self._bi_dir_rnn()[1] for _ in range(self.num_layers)], state_is_tuple=True)
            outputs, _ = tf.nn.bidirectional_dynamic_rnn(cell_fw, cell_bw, embedding_chars,
                                                         dtype=tf.float32)
            outputs = tf.concat(outputs, axis=2)
        return outputs

当代码修改成下面这样，则问题解决了，现在可以进行训练了