metadata
dataset_info:
features:
- name: translation
struct:
- name: bm
dtype: string
- name: fr
dtype: string
splits:
- name: train
num_bytes: 5543900
num_examples: 35070
download_size: 3496721
dataset_size: 5543900
configs:
- config_name: default
data_files:
- split: train
path: data/train-*