natural-questions
natural-questions copied to clipboard
Natural Questions (NQ) contains real user questions issued to Google search, and answers found from Wikipedia by annotators. NQ is designed for the training and evaluation of automatic question answer...
Hi, I've stumbled on a strange problem reading the files downloaded with `gsutil -m cp -R gs://natural_questions/v1.0 `. I try to find the answers using the `start_byte` and `end_byte` positions...
Hello everyone I tried downloading the data as explained in this link: [https://ai.google.com/research/NaturalQuestions/download](url) I keep getting the message: `gsutil requires python 2.7.` However i already have installed python2.7. I tried...
page not found for the link 'https://github.com/google-research-datasets/natural-questions/blob/master/data_utils.py' which apper in the Readme.md. And I also couldn't find the file in the repo.
Hi, I have some problems making a submission. I follow the FAQ in official site to send an email to [[email protected]](mailto:[email protected]) I wait for days but receive no response. So...
This is my code to open the data: jsonfilename = "/content/v1.0-simplified_nq-dev-all.jsonl.gz" with gzip.open(jsonfilename, 'r') as fin: data = json.loads(fin.read().decode('utf-8')) However, I'm getting this error: EOFError Traceback (most recent call last)...
Fixing typo in question on line 1817 "who won the workd series of poker 2017?" -> "who won the world series of poker 2017?"
I want to convert the source code of the a wikipedia webpage into the format provided by you in the competition. For an example if we look at [this](https://en.wikipedia.org/wiki/High_Commission_of_South_Africa,_London) webpage,...
First, thank you for the amazing dataset. As described in the paper, the first step of the annotation process is to classify the question as good or bad. But I...
there are many answers include long answers and short answers, if I train a model, which answer should I use, it couldn't use all of the condidates long answers, if...