metadata
dataset_info:
features:
- name: query
dtype: string
- name: product_id
dtype: uint64
- name: position
dtype: int64
- name: title
dtype: string
- name: pair_id
dtype: string
- name: score_linear
dtype: int64
- name: query_id
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 722522385
num_examples: 3803508
download_size: 310700853
dataset_size: 722522385
configs:
- config_name: default
data_files:
- split: train
path: data/train-*