---
dataset_info:
  features:
  - name: gt_blocks
    dtype: string
  - name: img
    dtype: image
  - name: page_bbox
    sequence: float32
  - name: classification
    dtype: string
  - name: language
    dtype: string
  - name: pdf
    dtype: binary
  - name: uuid
    dtype: int64
  splits:
  - name: train
    num_bytes: 775967309.75
    num_examples: 2138
  download_size: 760821924
  dataset_size: 775967309.75
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---