Open nghiahhnguyen opened 4 years ago
Moreover, I wondered if you still keep the question-document pair that was retrieved by the bm25 method, which was mentioned in Section 3.2 about the Answer Retriever in your paper: A Neural Network-based Framework for Non-factoid Question Answering. If you do not keep this information, would you mind giving me some instructions on how to retrieve the information myself using your code here? I would be grateful if you could provide me with some assistance.
hi, How can I run this code?
I have read about the FiQA dataset. However, I do not fully understand the meaning of your Pickle files. I understand the two files processed_question.pkl and processed_answers.pkl. However, I am not sure what dev_candidates.pkl include?