TypeError: can't pickle _thread.RLock objects

[middle-plat-ai]

Traceback (most recent call last): File "headline.py", line 255, in tf.app.run() File "/home/lyy/anaconda3/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 125, in run _sys.exit(main(argv)) File "headline.py", line 252, in main train() File "headline.py", line 173, in train model = create_model(sess, False) File "headline.py", line 147, in create_model forward_only=forward_only) File "/data/NLP_projects/textsum_projects/seq2seq-chinese-textsum/seq2seq_model.py", line 180, in init softmax_loss_function=softmax_loss_function) File "/home/lyy/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/legacy_seq2seq/python/ops/seq2seq.py", line 1215, in model_with_buckets decoder_inputs[:bucket[1]]) File "/data/NLP_projects/textsum_projects/seq2seq-chinese-textsum/seq2seq_model.py", line 179, in lambda x, y: seq2seq_f(x, y, False), File "/data/NLP_projects/textsum_projects/seq2seq-chinese-textsum/seq2seq_model.py", line 140, in seq2seq_f dtype=dtype) File "/home/lyy/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/legacy_seq2seq/python/ops/seq2seq.py", line 857, in embedding_attention_seq2seq encoder_cell = copy.deepcopy(cell) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 161, in deepcopy y = copier(memo) File "/home/lyy/anaconda3/lib/python3.6/site-packages/tensorflow/python/layers/base.py", line 385, in deepcopy setattr(result, k, copy.deepcopy(v, memo)) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 150, in deepcopy y = copier(x, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 215, in _deepcopy_list append(deepcopy(a, memo)) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 180, in deepcopy y = _reconstruct(x, memo, *rv) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 280, in _reconstruct state = deepcopy(state, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 150, in deepcopy y = copier(x, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 240, in _deepcopy_dict y[deepcopy(key, memo)] = deepcopy(value, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 180, in deepcopy y = _reconstruct(x, memo, *rv) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 280, in _reconstruct state = deepcopy(state, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 150, in deepcopy y = copier(x, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 240, in _deepcopy_dict y[deepcopy(key, memo)] = deepcopy(value, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 180, in deepcopy y = _reconstruct(x, memo, *rv) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 280, in _reconstruct state = deepcopy(state, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 150, in deepcopy y = copier(x, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 240, in _deepcopy_dict y[deepcopy(key, memo)] = deepcopy(value, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 180, in deepcopy y = _reconstruct(x, memo, *rv) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 280, in _reconstruct state = deepcopy(state, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 150, in deepcopy y = copier(x, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 240, in _deepcopy_dict y[deepcopy(key, memo)] = deepcopy(value, memo) File "/home/lyy/anaconda3/lib/python3.6/copy.py", line 169, in deepcopy rv = reductor(4) TypeError: can't pickle _thread.RLock objects

[Finley1991]

tf0.10.0与之后的版本embedding_attention_seq2seq有些变化，1.10.0版本不会出现，但之后版本会有这个问题。 如果想用多个backuts，会调用module_with_buckets，重复调用会使deepcopy(cell) 产生这个错误 修改\yourpath\tensorflow\contrib\legacy_seq2seq\python\ops\seq2seq.py： 我出现了同样的错误，最后是修改了 packages/tensorflow/contrib/legacy_seq2seq/python/ops/seq2seq.py　这个文件， 最开始添加一行：import tensorflow as tf （找到程序报错位置）添加： setattr(tf.contrib.rnn.GRUCell, 'deepcopy', lambda self, _: self) setattr(tf.contrib.rnn.BasicLSTMCell, 'deepcopy', lambda self, _: self) setattr(tf.contrib.rnn.MultiRNNCell, 'deepcopy', lambda self, _: self) 参考网址：https://blog.csdn.net/thormas1996/article/details/80743254

[minda163]

遇到类似的问题，把tensorflow改为1.10版本也没有用，还有其他的思路没有啊？？

[Finley1991]

修改\yourpath\tensorflow\contrib\legacy_seq2seq\python\ops\seq2seq.py：

setattr(tf.contrib.rnn.GRUCell, 'deepcopy', lambda self, _: self) setattr(tf.contrib.rnn.BasicLSTMCell, 'deepcopy', lambda self, _: self) setattr(tf.contrib.rnn.MultiRNNCell, 'deepcopy', lambda self, _: self)

我也找了很多，发现这个方法对我有效， 希望对你有帮助。 附上问题解决方案出处： https://blog.csdn.net/thormas1996/article/details/80743254

| | 王亚钒 | | [email protected] | 签名由网易邮箱大师定制 On 3/7/2019 15:19，minda163[email protected] wrote：

遇到类似的问题，把tensorflow改为1.10版本也没有用，还有其他的思路没有啊？？

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.