metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 7289367
num_examples: 24844
- name: test
num_bytes: 904221
num_examples: 3105
- name: validation
num_bytes: 914239
num_examples: 3106
download_size: 5263005
dataset_size: 9107827
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*