cmteb_retrieval / README.md
BeastyZ's picture
add language (#2)
e6a4d2e verified
metadata
license: apache-2.0
dataset_info:
  - config_name: cmedqa2
    features:
      - name: query
        dtype: string
      - name: positive
        sequence: string
      - name: negative
        sequence: string
      - name: answers
        sequence: 'null'
    splits:
      - name: train
        num_bytes: 1587455490
        num_examples: 100000
    download_size: 1027804069
    dataset_size: 1587455490
  - config_name: dureader
    features:
      - name: query
        dtype: string
      - name: positive
        sequence: string
      - name: negative
        sequence: string
      - name: answers
        sequence: 'null'
    splits:
      - name: train
        num_bytes: 7895977861
        num_examples: 86395
    download_size: 5019668526
    dataset_size: 7895977861
  - config_name: mmarco_merged
    features:
      - name: query
        dtype: string
      - name: positive
        sequence: string
      - name: negative
        sequence: string
      - name: answers
        sequence: 'null'
    splits:
      - name: train
        num_bytes: 24887177062
        num_examples: 388596
    download_size: 7142801140
    dataset_size: 24887177062
  - config_name: multi-cpr-ecom
    features:
      - name: query
        dtype: string
      - name: positive
        sequence: string
      - name: negative
        sequence: string
      - name: answers
        sequence: 'null'
    splits:
      - name: train
        num_bytes: 1778251126
        num_examples: 100000
    download_size: 1049289853
    dataset_size: 1778251126
  - config_name: multi-cpr-medical
    features:
      - name: query
        dtype: string
      - name: positive
        sequence: string
      - name: negative
        sequence: string
      - name: answers
        sequence: 'null'
    splits:
      - name: train
        num_bytes: 6924807931
        num_examples: 99999
    download_size: 3710282294
    dataset_size: 6924807931
  - config_name: multi-cpr-video
    features:
      - name: query
        dtype: string
      - name: positive
        sequence: string
      - name: negative
        sequence: string
      - name: answers
        sequence: 'null'
    splits:
      - name: train
        num_bytes: 1803174179
        num_examples: 100000
    download_size: 1290090817
    dataset_size: 1803174179
  - config_name: t2ranking
    features:
      - name: query
        dtype: string
      - name: positive
        sequence: string
      - name: negative
        sequence: string
      - name: answers
        sequence: 'null'
    splits:
      - name: train
        num_bytes: 531938618
        num_examples: 200376
    download_size: 344954364
    dataset_size: 531938618
configs:
  - config_name: cmedqa2
    data_files:
      - split: train
        path: cmedqa2/train-*
  - config_name: dureader
    data_files:
      - split: train
        path: dureader/train-*
  - config_name: mmarco_merged
    data_files:
      - split: train
        path: mmarco_merged/train-*
  - config_name: multi-cpr-ecom
    data_files:
      - split: train
        path: multi-cpr-ecom/train-*
  - config_name: multi-cpr-medical
    data_files:
      - split: train
        path: multi-cpr-medical/train-*
  - config_name: multi-cpr-video
    data_files:
      - split: train
        path: multi-cpr-video/train-*
  - config_name: t2ranking
    data_files:
      - split: train
        path: t2ranking/train-*
language:
  - zh