Neural network 运行两个RNN的Tensorflow：变量hidden/RNN/LSTMCell/W_0已存在

我试图同时运行两个rnn，并通过为每个

rnn\u cell.LSTMCell

定义两个变量范围，将它们的输出连接在一起。为什么我收到这个变量已经存在错误

ValueError:隐藏的变量/RNN/LSTMCell/W_0已存在， 不允许。您的意思是在VarScope中设置reuse=True吗

为什么它是“隐藏的/RNN/LSTMCell/W_0”而不是“隐藏的/向前的\u lstm_单元/RNN/LSTMCell/W_0”

---

只需使用

tf.variable\u scope

而不是

tf.name\u scope

tf.name\u scope

不会为使用tf.get\u variable（）创建的变量添加前缀

只需使用

tf.variable\u scope

而不是

tf.name\u scope

tf.name\u scope

不会向使用tf.get\u variable（）创建的变量添加前缀

    with tf.variable_scope('hidden', reuse=reuse): #reuse=None during training
        with tf.variable_scope('forward_lstm_cell'):
            lstm_fw_cell = tf.nn.rnn_cell.LSTMCell(num_units=self.num_hidden, use_peepholes=False, 
                                                initializer=tf.random_uniform_initializer(-0.003, 0.003),
                                                state_is_tuple=True)
            if not reuse:
                lstm_fw_cell = tf.nn.rnn_cell.DropoutWrapper(cell=lstm_fw_cell, input_keep_prob=0.7)

        with tf.variable_scope('backward_lstm_cell'):
            lstm_bw_cell = tf.nn.rnn_cell.LSTMCell(num_units=self.num_hidden, use_peepholes=False, 
                                                forget_bias=0.0, 
                                                initializer=tf.random_uniform_initializer(-0.003, 0.003),
                                                state_is_tuple=True)
            if not reuse:
                lstm_bw_cell = tf.nn.rnn_cell.DropoutWrapper(cell=lstm_bw_cell, input_keep_prob=0.7)

        with tf.name_scope("forward_lstm"):
            outputs_fw, output_states_fw  = tf.nn.dynamic_rnn(
                cell=lstm_fw_cell,
                inputs=embed_inputs_fw,
                dtype=tf.float32,
                sequence_length=self.seq_len_l
                )

        with tf.name_scope("backward_lstm"):
            outputs_bw, output_states_bw  = tf.nn.dynamic_rnn(
                cell=lstm_bw_cell,
                inputs=embed_inputs_bw,
                dtype=tf.float32,
                sequence_length=self.seq_len_r
                )