metadata
dataset_info:
features:
- name: descriptions
dtype: string
- name: labels
dtype: string
splits:
- name: train
num_bytes: 6245632
num_examples: 21009
download_size: 2869063
dataset_size: 6245632
configs:
- config_name: default
data_files:
- split: train
path: data/train-*