metadata
dataset_info:
features:
- name: text
dtype: string
- name: label
dtype: bool
- name: type
dtype: string
- name: relation
dtype: string
splits:
- name: train
num_bytes: 1948680
num_examples: 18162
download_size: 436095
dataset_size: 1948680
configs:
- config_name: default
data_files:
- split: train
path: data/train-*