metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: news_text
dtype: string
- name: title
dtype: string
- name: hyperpartisan
dtype: bool
- name: url
dtype: string
- name: published_at
dtype: string
- name: bias
dtype:
class_label:
names:
'0': right
'1': right-center
'2': least
'3': left-center
'4': left
- name: text
dtype: string
- name: uid
dtype: int64
splits:
- name: train
num_bytes: 5549889491
num_examples: 600000
- name: validation
num_bytes: 1906305570
num_examples: 150000
download_size: 4230482849
dataset_size: 7456195061
