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: 36773693
num_examples: 13871
download_size: 16334899
dataset_size: 36773693
configs:
- config_name: default
data_files:
- split: train
path: data/train-*