metadata
dataset_info:
features:
- name: translation
struct:
- name: de
dtype: string
- name: en
dtype: string
splits:
- name: train
num_bytes: 41493189
num_examples: 206112
- name: validation
num_bytes: 206648
num_examples: 888
- name: test
num_bytes: 1576722
num_examples: 8079
download_size: 26881776
dataset_size: 43276559
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*