---
dataset_info:
  features:
  - name: qid
    dtype: string
  - name: question
    dtype: string
  - name: images
    sequence: image
  - name: answer
    dtype: string
  - name: question_info
    dtype: string
  - name: type
    dtype: string
  - name: choices
    sequence: string
  splits:
  - name: train
    num_bytes: 25689730.0
    num_examples: 532
  download_size: 21962402
  dataset_size: 25689730.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---