metadata
dataset_info:
features:
- name: bright_id
dtype: string
- name: bright_doc
dtype: string
- name: bright_split
dtype: string
- name: scandi_id
dtype: string
- name: scandi_url
dtype: string
- name: scandi_title
dtype: string
- name: scandi_text
dtype: string
- name: scandi_language
dtype: string
- name: scandi_dist
dtype: float64
splits:
- name: train
num_bytes: 348628207
num_examples: 5650
download_size: 105064858
dataset_size: 348628207
configs:
- config_name: default
data_files:
- split: train
path: data/train-*