metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: label
dtype:
class_label:
names:
'0': negative
'1': neutral
'2': positive
splits:
- name: train
num_bytes: 677419
num_examples: 4999
download_size: 378755
dataset_size: 677419
configs:
- config_name: default
data_files:
- split: train
path: data/train-*