Lots-of-LoRAs/task1239_ted_translation_gl_ja

task1239_ted_translation_gl_ja 数据集是一个文本生成任务的数据集，由众包方式创建，包含英文到日语的TED演讲翻译。数据集分为训练集、验证集和测试集，共有5112个训练样本和1278个验证与测试样本。数据集的配置名称为 plain_text，包含输入文本、输出文本和样本ID三个字段。此数据集适用于研究机器翻译、文本生成等自然语言处理任务。

The task1239_ted_translation_gl_ja dataset is a text generation task dataset created through crowdsourcing, containing English to Japanese translations of TED talks. The dataset is divided into training, validation, and test sets, with a total of 5112 training samples and 1278 validation and test samples. The configuration name of the dataset is plain_text, which includes three fields: input text, output text, and sample ID. This dataset is suitable for researching machine translation, text generation, and other natural language processing tasks.