metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: unsandhied
dtype: string
splits:
- name: train
num_bytes: 10973642
num_examples: 99889
- name: validation
num_bytes: 470141
num_examples: 4200
- name: test
num_bytes: 470141
num_examples: 4200
- name: test_500
num_bytes: 58711
num_examples: 500
- name: validation_500
num_bytes: 58795
num_examples: 500
download_size: 7465295
dataset_size: 12031430
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*
- split: test_500
path: data/test_500-*
- split: validation_500
path: data/validation_500-*