metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: antedecent_whole
dtype: string
- name: antedecent_root
dtype: string
- name: anaphora
dtype: string
- name: category
dtype: string
- name: sempos
dtype: string
- name: corpus
dtype: string
splits:
- name: train
num_bytes: 10569880
num_examples: 33667
- name: validation
num_bytes: 1357056
num_examples: 4208
- name: test
num_bytes: 1337257
num_examples: 4216
download_size: 8260824
dataset_size: 13264193
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*