doctor3030/newsboard-news-categories

该数据集包含多个特征，如标题（header）、正文（body）、类别（categories）、所有类别（all_categories）、文本（text）和标签（labels）。all_categories特征包含24个不同的类别，如法国、澳大利亚、墨西哥等。数据集分为训练集和测试集，分别包含47236和11809个样本。

The dataset includes multiple features such as header, body, categories, all_categories, text, and labels. Among them, header and body are string types, categories and text are also string types, while all_categories is a categorical label with multiple classes, and labels are sequences of float32. The dataset is divided into a training set and a test set, containing 47236 and 11809 samples respectively. The download size of the dataset is 291405881 bytes, and the total size is 485864229.0 bytes.