---
dataset_info:
  config_name: train
  features:
  - name: image
    dtype: image
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: split
    dtype: string
  splits:
  - name: train
    num_bytes: 45709895.0
    num_examples: 400
  download_size: 45599301
  dataset_size: 45709895.0
configs:
- config_name: train
  data_files:
  - split: train
    path: train/train-*
---