metadata
dataset_info:
features:
- name: url
dtype: string
- name: pdf_file
dtype: string
- name: page_number
dtype: int64
- name: text
dtype: string
splits:
- name: train
num_bytes: 38029949
num_examples: 15751
download_size: 17000220
dataset_size: 38029949
configs:
- config_name: default
data_files:
- split: train
path: data/train-*