---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: image_url
    dtype: string
  - name: query
    dtype: string
  - name: answer
    dtype: string
  - name: choice
    dtype: string
  - name: question_type
    dtype: string
  splits:
  - name: train
    num_bytes: 372978
    num_examples: 850
  download_size: 144695
  dataset_size: 372978
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---