spacemanidol/MSMARCO

Issues

• Regarding the Test Set for Q&A

  #38 opened 5 years ago by tahmedge
  2

• dev_as_references.json is from V1?

  #37 opened 5 years ago by deeplearningmachine
  2

• OpenKPAnnotations.tsv for Key Phrase Extraction

  #36 opened 5 years ago by zeynepakkalyoncu
  1

• Invalid line breaks in the top1000 TSV files of the reranking datasets

  #31 opened 5 years ago by ikuyamada
  4

• Passage IDs in Qna Dataset

  #34 opened 5 years ago by spacemanidol
  1

• Partially duplicated passages extracted

  #33 opened 5 years ago by daltonj
  4

• [encoding,Â] top1000.dev.tsv

  #26 opened 6 years ago by Albert-Ma
  6

• Different number of queries in collectionandqueries.tar.gz and top1000.dev.tar.gz

  #29 opened 6 years ago by rodrigonogueira4
  1

• KeyError for converttowellformed.py

  #28 opened 6 years ago
  1

• Need more explanation about Reranking Dataset

  #27 opened 6 years ago by Planck35
  1

• encoding, top1000.train, qrels.train

  #25 opened 6 years ago by linxihui
  1

• Collection paragraph metadata

  #19 opened 6 years ago by daltonj
  3

• How to understand the gain score about 'No Answer Present' ?

  #1 opened 7 years ago by jellying
  1

• some bugs about the code

  #2 opened 6 years ago by javacjh
  4

• there are many bugs in the evaluation script

  #3 opened 6 years ago by xinyadu
  1

• Problem with Utils/tojson.py

  #6 opened 6 years ago by xycforgithub
  1

• No module named mrcqa.modules

  #8 opened 6 years ago by wshinigamic
  1

• [docs] Data is not JSONL

  #9 opened 6 years ago by juharris
  1

• Uncommon train / dev / test split of ranking dataset

  #11 opened 6 years ago by drennings
  5

• Include statistics on ranking dataset in documentation

  #12 opened 6 years ago by drennings
  2

• MSMARCOV2/Ranking/README.md is not formatted correctly

  #13 opened 6 years ago by rodrigonogueira4
  1

• How were passage reranking triples generated?

  #14 opened 6 years ago by chsasank
  2

• keyerror

  #16 opened 6 years ago by zkt12
  3

• Cannot find the top1000.eval for testing

  #17 opened 6 years ago by Qiaoyf96
  1

• Full Document May be incorrect tokenization in document_text

  #18 opened 6 years ago by daltonj
  2

• BM25 relevance values for top 1000 eval/dev?

  #21 opened 6 years ago by amirj
  1

• Training data with QID and PID

  #22 opened 6 years ago by QingyaoAi
  2

• Broken eval script link in Ranking/README.md file

  #23 opened 6 years ago by rodrigonogueira4
  1