metadata
dataset_info:
features:
- name: sample_id
dtype: string
- name: source_text
dtype: string
- name: source_language
dtype: string
- name: target_language
dtype: string
- name: domain
dtype: string
- name: google_comparable
dtype: bool
splits:
- name: train
num_bytes: 2115351
num_examples: 21000
download_size: 592628
dataset_size: 2115351
configs:
- config_name: default
data_files:
- split: train
path: data/train-*