Ebenezer Scrooge

I have to rewrite the model with Keras...

The spot where the issue occurs: def basic_tokenizer(sentence): """Very basic tokenizer: split the sentence into a list of tokens.""" words = [] #sentence = tf.compat.as_bytes(sentence) for space_separated_fragment in sentence.strip().split(): words.extend(re.split(_WORD_SPLIT,...

Runtime Environment: Ubuntu 16.04 Python 3.5 TensorFlow 1.5.0 PyCharm Community data were downloaded from the link of baidunetdisk given in README.MD

I found a very similar repo about SeqGAN: https://github.com/vpegasus/seqGan_chatbot/blob/master/utils/data_utils.py and from his code I found the inexistence of regular expression part of this function: def basic_tokenizer(sentence): """Very basic tokenizer: split...

我这也是空的，怎么回事？

https://github.com/AlucardNosferatu/chatbot/blob/master/seqGanChatbot/gen_data/test.ids2500.answer https://raw.githubusercontent.com/AlucardNosferatu/chatbot/master/seqGanChatbot/gen_data/train.ids2500.answer 这里面看无论test还是train的id都是3，感觉词语矢量化是不是有些问题。。。？

我的词典文件和这个#56 情况一样 不知道为啥是空的

> I have solved this problem，try to use the lower version. Which version? I was using 8.2.4

> > > I have solved this problem，try to use the lower version. > > > > > > Which version? I was using 8.2.4 > > Use python 3.6，SWI-Prolog...

> python 3.7 may be not supported. [https://github.com/cython/cython/issues/1955](url)