---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: id
    dtype: int64
  splits:
  - name: train
    num_bytes: 430646390.0
    num_examples: 635
  - name: test
    num_bytes: 173859026.0
    num_examples: 250
  download_size: 589640106
  dataset_size: 604505416.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---