metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 15406598
num_examples: 29003
- name: validation
num_bytes: 889805
num_examples: 1710
- name: test
num_bytes: 853238
num_examples: 1581
download_size: 6389333
dataset_size: 17149641
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*