---
dataset_info:
  features:
  - name: original_caption
    dtype: string
  - name: image_url
    dtype: string
  - name: generated_qa
    dtype: string
  splits:
  - name: train
    num_bytes: 3897
    num_examples: 5
  download_size: 8354
  dataset_size: 3897
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---