README.md

metadata

language:
  - vi
dataset_info:
  features:
    - name: sentence1
      dtype: string
    - name: sentence2
      dtype: string
    - name: score
      dtype: float64
    - name: similarity
      dtype: float64
    - name: type
      dtype: string
    - name: hard
      dtype: int64
  splits:
    - name: train
      num_bytes: 31051336186.85252
      num_examples: 11052543
    - name: test
      num_bytes: 112377164.9179832
      num_examples: 40000
    - name: eval
      num_bytes: 28094291.2294958
      num_examples: 10000
  download_size: 17939520772
  dataset_size: 31191807643
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
      - split: eval
        path: data/eval-*