---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 348211463.208
    num_examples: 19768
  download_size: 328827117
  dataset_size: 348211463.208
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---