metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: unsandhied
dtype: string
splits:
- name: train
num_bytes: 21289168
num_examples: 49768
- name: validation
num_bytes: 440708
num_examples: 1909
- name: test
num_bytes: 444777
num_examples: 1910
download_size: 12753404
dataset_size: 22174653
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*