metadata
dataset_info:
features:
- name: query-id
dtype: string
- name: corpus-id
dtype: string
- name: score
dtype: int32
splits:
- name: test
num_bytes: 49658
num_examples: 2672
download_size: 17141
dataset_size: 49658
configs:
- config_name: default
data_files:
- split: test
path: data/test-*