Andrew

No, iepy can't process Chinese, so i didn't use it any more.

Yeah, I set the segment length parameter `L` to be 10 which is longer than the previous one and the neglogev is normal. But at template extraction phase another question...

Continue the above question. After I check the infer.py file and I am confused at line 71. if var jj iterates over L, we can't guarantee that the first dim...

@eromoe Right now, I want iepy to customize to Chinese, could you give me a hand ?

@eromoe All right. Thanks a lot. Now I was also met with this trouble, I really need someone could help me.

> Could you pls upload swda_data file, since the repo you mentioned in the previous issue does not have the load_file() function. Thx! It is used to load preprocessed data....

> On lines 234 and 235 you input the 'data' and 'labels', which you've referred to as virtual data. To train the model, I am assuming data should be a...

I am sorry data preprocessing file swda_data.py was removed by mistake. You can preprocess the original data and convert dialogues to the form like line #L250 and labels to the...

I used the dataset parsed by @cgpotts, please refer to his [repository](https://github.com/cgpotts/swda)

> Hi, what is the version of TensorFlow you used in your codes? Thx I guess 1.08. It’s been a long time now.