metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: antedecent_whole
dtype: string
- name: antedecent_root
dtype: string
- name: anaphora
dtype: string
- name: source_file
dtype: string
splits:
- name: train
num_bytes: 7467555
num_examples: 22822
download_size: 4193302
dataset_size: 7467555
configs:
- config_name: default
data_files:
- split: train
path: data/train-*