README.md

metadata

dataset_info:
  features:
    - name: clip_name
      dtype: string
    - name: human_caption
      dtype: string
  splits:
    - name: train
      num_bytes: 1544750
      num_examples: 500
  download_size: 806248
  dataset_size: 1544750
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*