metadata
dataset_info:
features:
- name: title
dtype: string
- name: date
dtype: string
- name: docname
dtype: string
- name: source_url
dtype: string
- name: pdf_url
dtype: string
- name: text
dtype: string
- name: num_chars
dtype: int64
splits:
- name: train
num_bytes: 2770186
num_examples: 374
download_size: 1160831
dataset_size: 2770186
configs:
- config_name: default
data_files:
- split: train
path: data/train-*