---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: question
    dtype: string
  - name: answer
    dtype: string
  splits:
  - name: train
    num_bytes: 306352877
    num_examples: 1599
  - name: val
    num_bytes: 76886375
    num_examples: 400
  download_size: 382668952
  dataset_size: 383239252
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: val
    path: data/val-*
---