---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    sequence: string
  - name: index
    dtype: int64
  splits:
  - name: train
    num_bytes: 51733249508.27
    num_examples: 81355
  download_size: 51732268420
  dataset_size: 51733249508.27
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---