metadata
dataset_info:
features:
- name: text
dtype: string
- name: label
dtype: int64
- name: sa
dtype: int64
splits:
- name: train
num_bytes: 3843951
num_examples: 50000
- name: test
num_bytes: 154027
num_examples: 2000
- name: dev
num_bytes: 152283
num_examples: 2000
download_size: 2169706
dataset_size: 4150261
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: dev
path: data/dev-*