metadata
dataset_info:
features:
- name: document_id
dtype: string
- name: corpus
dtype: string
- name: sentence_ls
list: string
- name: metadata
struct:
- name: title
dtype: string
- name: author
dtype: string
- name: publisher
dtype: string
- name: date
dtype: string
- name: topic
dtype: string
- name: original_topic
dtype: string
- name: category
dtype: string
splits:
- name: train
num_bytes: 4621522905
num_examples: 729280
download_size: 2651908619
dataset_size: 4621522905
configs:
- config_name: default
data_files:
- split: train
path: data/train-*