metadata
dataset_info:
features:
- name: subject
dtype: string
- name: descriptor
dtype: string
splits:
- name: train
num_bytes: 182700
num_examples: 13050
download_size: 0
dataset_size: 182700
configs:
- config_name: default
data_files:
- split: train
path: data/train-*