metadata
dataset_info:
features:
- name: user_id
dtype: string
- name: user_profile
dtype: string
- name: star_gap
dtype: float64
- name: chosen
struct:
- name: business_id
list: string
- name: business_name
list: string
- name: rating
list: float64
- name: rejected
struct:
- name: business_id
list: string
- name: business_name
list: string
- name: rating
list: float64
- name: rating_map
dtype: string
- name: business_profiles
list: string
splits:
- name: train
num_bytes: 79096399
num_examples: 13943
download_size: 39319959
dataset_size: 79096399
configs:
- config_name: default
data_files:
- split: train
path: data/train-*