metadata
dataset_info:
features:
- name: topic
dtype: string
- name: syntagm
dtype: string
- name: labels
dtype: int64
- name: root
dtype: string
- name: premise
dtype: string
- name: hypothesis
dtype: string
splits:
- name: train
num_bytes: 5319468
num_examples: 34295
- name: test
num_bytes: 1134796
num_examples: 7350
- name: eval
num_bytes: 1139042
num_examples: 7349
download_size: 2212186
dataset_size: 7593306
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: eval
path: data/eval-*