metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 6697049
num_examples: 24868
- name: test
num_bytes: 841002
num_examples: 3109
- name: validation
num_bytes: 833880
num_examples: 3109
download_size: 5086746
dataset_size: 8371931
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*