metadata
dataset_info:
features:
- name: translation
struct:
- name: bm
dtype: string
- name: fr
dtype: string
splits:
- name: train
num_bytes: 168300686
num_examples: 700070
download_size: 108388335
dataset_size: 168300686
configs:
- config_name: default
data_files:
- split: train
path: data/train-*