---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: question
    dtype: string
  - name: answer
    dtype: string
  splits:
  - name: train
    num_bytes: 1074552713.904604
    num_examples: 1928
  - name: test
    num_bytes: 269195519.0953961
    num_examples: 483
  download_size: 1343698644
  dataset_size: 1343748233.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---