---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: image_filename
    dtype: string
  - name: query
    dtype: string
  - name: answer
    dtype: string
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 2695948104.0
    num_examples: 3000
  - name: test
    num_bytes: 2239357646.156
    num_examples: 2511
  download_size: 4913465171
  dataset_size: 4935305750.156
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---