---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: answer
    dtype: string
  - name: query
    dtype: string
  splits:
  - name: train
    num_bytes: 2935563048.46
    num_examples: 552340
  download_size: 2780097738
  dataset_size: 2935563048.46
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---