metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 29255592
num_examples: 76052
- name: validation
num_bytes: 2685837
num_examples: 7069
- name: test
num_bytes: 2563727
num_examples: 6740
download_size: 9785153
dataset_size: 34505156
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*