metadata
dataset_info:
features:
- name: query-id
dtype: string
- name: corpus-id
dtype: string
- name: score
dtype: int32
splits:
- name: train
num_bytes: 75348
num_examples: 996
download_size: 17729
dataset_size: 75348
configs:
- config_name: default
data_files:
- split: train
path: data/train-*