zhhhzhang

ERROR:pytorch_transformers.tokenization_utils:Couldn't reach server to download vocabulary.

2

how to do?

Thanks for your sharing, but I have a question that how to use it ?

2

how to use it ?

关于data_pro.py的若干问题

8

作者你好！ 首先感谢您的分享！我在别的数据集运行了data_pro.py，比如reviews_Amazon_Instant_Video_data，报错了，有几个问题想请教一下作者 1. index = data_test.index[data_test['item_id'] == iid].tolist()[0] 这个地方为什么只取索引[0]？这样处理并没有真正把所有的测试集都去掉，之去掉了相同item_id的第一个 2. data_train = pd.concat([data_train, data_test.iloc[uid_concat_index]]) ，iloc是按照位置进行查找的，我看了作者的代码应该想按照索引index进行删除，此处会报错，是不是应该替换成loc? 3. 作者换份train test和val，但是并没有用val，不知道是怎么考虑的？ 谢谢！

two more question?

1

thanks for you code! and two more question: 1. you said you will release a newer version in the first half of 2020, i wonder what is the time? 2....

question about dataset?

2

There are some files which are missed in the data.  Could you please provide these files so that I can follow your step? Thanks very much.

请教数据集问题？

1

首先感谢提供数据集和相关代码！！！ 想请问一下作者能否提供一下原始的数据集？而不是预处理为bert之后的格式！ 谢谢！

我跑了这个代码我发现啤酒-尿不湿 和尿不湿-啤酒都统计了，难道和顺序有关？

i wonder if the task of implement is for text summarization? could you share a toy_dataset so that i can make my own dataset? just an data example?

请问作者关于注入法律知识的continual training与有监督的指令微调两种训练方式有什么区别？

11

如题！两种训练方式的训练目标是一样的么？两者的训练数据构造有什么区别？谢谢！