google-research-datasets / natural-questions

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 answering systems.

google-research-datasets/natural-questions Issues

pre-processing code for document_html to document_tokens
Closed 4 years ago1
Encoding issue in traindata
Updated a year ago2
Download data
Updated 2 years ago2
Don't have access to 'data_util.py' in Readme
Updated 2 years ago
cannot reach out to official email
Updated 2 years ago
Prediction on Custom Text
Closed 3 years ago
EOFError: Compressed file ended before the end-of-stream marker was reached (on Google Colab)?
Updated 3 years ago
How to get "Long Answer Candidates" from wikipedia source code.
Updated 4 years ago
Running nq_browser.py on Windows OS
Closed 4 years ago4
Question identification: good or bad?
Updated 4 years ago1
Python3 data processing
Closed 5 years ago1
Regarding Char offset
Closed 5 years ago1
Running simplify_nq_data on dev test?
Closed 5 years ago
how to train with this data
Updated 5 years ago1
why simplified test data just has token start and end, doesn' have byte start and byte end
Updated 5 years ago1
clarification of 'score'
Updated 5 years ago1
where is baseline of bert?
Updated 5 years ago2
Can anybody explain implementation level details of "closest question" baseline, as our results are quite different from the results given in Natural Questions Research paper.
Updated 5 years ago
Convert token to text
Updated 5 years ago2
Error when loading gold file
Closed 6 years ago2