---
dataset_info:
  features:
  - name: questionId
    dtype: int64
  - name: question
    dtype: string
  - name: answers
    sequence: string
  - name: image
    dtype: image
  splits:
  - name: train
    num_bytes: 10097474383.96932
    num_examples: 14514
  download_size: 1836713139
  dataset_size: 10097474383.96932
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---