---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: paper_id
    dtype: string
  - name: page_number
    dtype: int64
  - name: image_hash
    dtype: string
  - name: source_tar
    dtype: string
  - name: source_yymm
    dtype: string
  - name: total_pages
    dtype: int64
  - name: image_filename
    dtype: string
  splits:
  - name: train
    num_bytes: 242019414.0
    num_examples: 1000
  download_size: 241696882
  dataset_size: 242019414.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---