README.md

---
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: 1485376783.75
    num_examples: 10074
  download_size: 302998294
  dataset_size: 1485376783.75
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
task_categories:
- document-question-answering
- visual-document-retrieval
---