metadata
dataset_info:
features:
- name: id
dtype: int64
- name: language
dtype: string
- name: target_text
dtype: string
- name: back_translations
dtype: string
- name: reference_text
dtype: string
- name: __index_level_0__
dtype: int64
- name: bert_score_f1
dtype: float64
- name: bert_score_precision
dtype: float64
- name: bert_score_recall
dtype: float64
splits:
- name: train
num_bytes: 235536539
num_examples: 558128
download_size: 132561139
dataset_size: 235536539
configs:
- config_name: default
data_files:
- split: train
path: data/train-*