metadata
dataset_info:
features:
- name: src_lang
dtype: string
- name: tgt_lang
dtype: string
- name: src_text
dtype: string
- name: tgt_text
dtype: string
splits:
- name: validation
num_bytes: 4652553
num_examples: 30000
- name: test_data
num_bytes: 7319697
num_examples: 42000
- name: train
num_bytes: 3903276358
num_examples: 28902212
download_size: 5967652250
dataset_size: 3915248608
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test_data
path: data/test_data-*