---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: id
    dtype: int64
  splits:
  - name: train
    num_bytes: 283068596.0
    num_examples: 792
  - name: test
    num_bytes: 90920952.0
    num_examples: 250
  download_size: 365551655
  dataset_size: 373989548.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---