metadata
dataset_info:
features:
- name: query
dtype: string
- name: alternative
dtype: string
- name: semantic_score
dtype: float64
- name: num_query
dtype: int64
- name: num_alternative
dtype: int64
splits:
- name: train
num_bytes: 10741285
num_examples: 107917
- name: test
num_bytes: 1193976
num_examples: 11991
download_size: 7527827
dataset_size: 11935261
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*