---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: cropped_image
    dtype: image
  - name: answer
    dtype: string
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 408334081.2011177
    num_examples: 26041
  download_size: 408093962
  dataset_size: 408334081.2011177
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---