metadata
dataset_info:
features:
- name: text
dtype: string
- name: predicted_class
dtype: string
- name: confidence
dtype: float16
splits:
- name: train
num_bytes: 15294599
num_examples: 16000
download_size: 7580515
dataset_size: 15294599
configs:
- config_name: default
data_files:
- split: train
path: data/train-*