README.md

metadata

dataset_info:
  - config_name: af-en
    features:
      - name: src_doc_id
        dtype: string
      - name: tgt_doc_id
        dtype: string
      - name: lang_pair
        dtype: string
      - name: src_doc
        struct:
          - name: ids
            sequence: string
          - name: sentences
            sequence: string
      - name: tgt_doc
        struct:
          - name: ids
            sequence: string
          - name: sentences
            sequence: string
      - name: alignment
        list:
          - name: src
            list: string
          - name: tgt
            list: string
          - name: aligner-score
            dtype: float32
          - name: bicleaner-score
            dtype: float32
          - name: bifixer-score
            dtype: float32
    splits:
      - name: train
        num_bytes: 22358649843
        num_examples: 1121166
    download_size: 6420749225
    dataset_size: 22358649843
  - config_name: az-en
    features:
      - name: src_doc_id
        dtype: string
      - name: tgt_doc_id
        dtype: string
      - name: lang_pair
        dtype: string
      - name: src_doc
        struct:
          - name: ids
            sequence: string
          - name: sentences
            sequence: string
      - name: tgt_doc
        struct:
          - name: ids
            sequence: string
          - name: sentences
            sequence: string
      - name: alignment
        list:
          - name: src
            list: string
          - name: tgt
            list: string
          - name: aligner-score
            dtype: float32
          - name: bicleaner-score
            dtype: float32
          - name: bifixer-score
            dtype: float32
    splits:
      - name: train
        num_bytes: 9761021558
        num_examples: 732657
    download_size: 3230027128
    dataset_size: 9761021558
configs:
  - config_name: af-en
    data_files:
      - split: train
        path: af-en/train-*
  - config_name: az-en
    data_files:
      - split: train
        path: az-en/train-*