metadata
dataset_info:
features:
- name: id
dtype: string
- name: version
dtype: string
- name: query
dtype: string
- name: doc1
dtype: string
- name: doc2
dtype: string
- name: label
dtype: int64
splits:
- name: train
num_bytes: 686764
num_examples: 450
download_size: 173613
dataset_size: 686764
configs:
- config_name: default
data_files:
- split: train
path: data/train-*