speedcell4/opus-m2m

该数据集包含两对文本序列和它们对应的语言，以及文本的大小信息。数据集分为训练集、验证集、测试集和一个名为zero的集合，每个集合包含不同数量的示例。提供了默认配置文件，指明了各个集合的数据文件路径。

The dataset includes two pairs of text sequences and their corresponding languages, along with the size information of the texts. It is divided into training set, validation set, test set, and a set named zero, each containing a different number of examples. A default configuration file is provided, indicating the data file paths for each set.