metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 7178882
num_examples: 28465
- name: validation
num_bytes: 811396
num_examples: 3558
- name: test
num_bytes: 823679
num_examples: 3559
download_size: 4885600
dataset_size: 8813957
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*