metadata
dataset_info:
features:
- name: premise
dtype: string
- name: hypothesis
dtype: string
- name: category
dtype: string
- name: label
dtype:
class_label:
names:
'0': NotEntail
'1': Entail_or_Paraphrase
splits:
- name: train
num_bytes: 99987
num_examples: 300
- name: validation
num_bytes: 18037
num_examples: 50
- name: test
num_bytes: 32605
num_examples: 100
download_size: 98189
dataset_size: 150629
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*