metadata
dataset_info:
features:
- name: context
dtype: string
- name: answerA
dtype: string
- name: answerB
dtype: string
- name: label
dtype:
class_label:
names:
'0': answerA
'1': answerB
splits:
- name: train
num_bytes: 1231148
num_examples: 9248
- name: validation
num_bytes: 167984
num_examples: 1267
download_size: 704014
dataset_size: 1399132
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*