metadata
dataset_info:
features:
- name: translation
dtype:
translation:
languages:
- en
- id
splits:
- name: train
num_bytes: 78698973
num_examples: 1000000
- name: validation
num_bytes: 180024
num_examples: 2000
- name: test
num_bytes: 177685
num_examples: 2000
download_size: 57693678
dataset_size: 79056682
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*