---
dataset_info:
  features:
  - name: pdf
    dtype: binary
  - name: doc_id
    dtype: string
  splits:
  - name: train
    num_bytes: 14802127
    num_examples: 1
  download_size: 13103577
  dataset_size: 14802127
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---