metadata
dataset_info:
features:
- name: text
dtype: string
- name: category
dtype: string
- name: items
dtype: string
splits:
- name: train
num_bytes: 622317
num_examples: 2482
download_size: 204878
dataset_size: 622317
configs:
- config_name: default
data_files:
- split: train
path: data/train-*