metadata
dataset_info:
features:
- name: goods_nm
dtype: string
- name: category
dtype: string
- name: label
dtype: int64
splits:
- name: train
num_bytes: 316703987
num_examples: 3400000
download_size: 149786014
dataset_size: 316703987
configs:
- config_name: default
data_files:
- split: train
path: data/train-*