---
dataset_info:
  features:
    - name: raw_image_url
      dtype: string
    - name: text1
      dtype: string
    - name: text2
      dtype: string
    - name: text3
      dtype: string
    - name: syn_text
      dtype: string
  splits:
    - name: train
      num_bytes: 8729944000
      num_examples: 10000
  download_size: 8729944000
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
---