---
dataset_info:
  features:
  - name: key
    dtype: string
  - name: romanization
    dtype: string
  - name: hanzi
    dtype: string
  - name: script_tags
    list: string
  - name: printed_pages
    dtype: string
  - name: pdf_pages
    dtype: string
  - name: n_pages
    dtype: int32
  - name: page_index
    dtype: int32
  - name: image
    dtype: image
  splits:
  - name: train
    num_bytes: 195056701
    num_examples: 76
  download_size: 195062582
  dataset_size: 195056701
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---