metadata
dataset_info:
features:
- name: source
dtype: string
- name: domain
dtype: string
- name: sentence
dtype: string
- name: triples
dtype: string
splits:
- name: train
num_bytes: 5246148
num_examples: 16447
download_size: 2439985
dataset_size: 5246148
configs:
- config_name: default
data_files:
- split: train
path: data/train-*