---
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
dataset_info:
  features:
  - name: id
    dtype: string
  - name: image_0
    dtype: image
  - name: image_1
    dtype: image
  - name: question
    dtype: string
  splits:
  - name: train
    num_bytes: 12377693948
    num_examples: 15000
  download_size: 12377496522
  dataset_size: 12377693948
---