if001/wikimedia_ja_short

该数据集包含1,384,986个训练示例，总大小约为7.02 GB，下载大小约为3.64 GB。每个示例具有四个特征：id（字符串类型，唯一标识符）、url（字符串类型，来源链接）、title（字符串类型，标题）和text（字符串类型，文本内容）。数据集仅提供train分割，数据文件路径为data/train-*。由于README未提供具体描述，无法确定数据集的特定用途或领域，但根据特征推测可能用于文本分析或自然语言处理任务。

This dataset contains 1,384,986 training examples with a total size of approximately 7.02 GB and a download size of approximately 3.64 GB. Each example includes four features: id (string type, unique identifier), url (string type, source link), title (string type, title), and text (string type, text content). The dataset only provides a train split, with data files located at data/train-*. As the README does not include a specific description, the exact purpose or domain of the dataset cannot be determined, but based on the features, it is likely intended for text analysis or natural language processing tasks.