tensorflow报错解决方案：ValueError: Variable bidirectional_rnn/fw/lstm_cell/kernel already exists, disallow

最新推荐文章于 2024-07-10 09:58:45 发布

保护我方鲁班七号

最新推荐文章于 2024-07-10 09:58:45 发布

阅读量5.8k

点赞数 4

分类专栏：深度学习文章标签： tensorflow python 命名报错

本文链接：https://blog.csdn.net/m511655654/article/details/85125467

版权

深度学习专栏收录该内容

20 篇文章 1 订阅

订阅专栏

ValueError: Variable bidirectional_rnn/fw/lstm_cell/kernel already exists, disallowed. Did you mean to set reuse=True or reuse=tf.AUTO_REUSE in VarScope? Originally defined at:

File "XX.py", line 105, in rnn
outputs, _, _ = tf.contrib.rnn.static_bidirectional_rnn(cell_fw, cell_bw, inputs, dtype=tf.float32)
File "XX.py", line 57, in __init__
self.rnn()
File "XX.py", line 248, in <module>
model = TextRNN(config)

运行tensorflow程序时，如果出现以上类似的报错，是程序命名空间相同产生错误，对每个函数定义不同的name，即可解决。

cell_fw2 = tf.nn.rnn_cell.LSTMCell(self.config.hidden_dim, name='cell_fw_2')

cell_bw2 = tf.nn.rnn_cell.LSTMCell(self.config.hidden_dim, name='cell_bw_2')