Hi, how can i get the tokenized story? in newsqa-data-tokenized-v1.csv only have fields: question, answer_char_ranges , is_answer_absent,is_question_bad,story_text,answer_token_ranges ,sentence_starts.
Thanks for trying out NewsQA! Sorry about that, it should be better documented. Looking at the code, it looks like story_text has the tokens separated by spaces.
Hi, how can i get the tokenized story? in newsqa-data-tokenized-v1.csv only have fields: question, answer_char_ranges , is_answer_absent,is_question_bad,story_text,answer_token_ranges ,sentence_starts.
i need something like tokenized_story_text
Thanks