metadata
dataset_info:
features:
- name: mg
dtype: string
- name: fr
dtype: string
- name: token_mg
dtype: int64
- name: token_fr
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 86160016.36893491
num_examples: 412805
- name: test
num_bytes: 21540160.63106508
num_examples: 103202
download_size: 70935696
dataset_size: 107700177
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*