metadata
dataset_info:
features:
- name: text
dtype: string
- name: span
dtype: string
- name: label
dtype: string
- name: ordinal
dtype: int64
splits:
- name: train
num_bytes: 5221120
num_examples: 2653
download_size: 568993
dataset_size: 5221120
configs:
- config_name: default
data_files:
- split: train
path: data/train-*