quentintaranpino/image-moderation

该数据集包含图像、标签和来源三个主要特征。标签是一个序列，包含多个类别的标签，如仇恨、羞辱、骚扰、暴力、伤害、残忍、性内容、裸体内容、犯罪计划、武器或物质滥用、自残、动物虐待、灾难或紧急情况以及政治内容。数据集被分割为训练集，包含49,371个样本，总大小为2,185,425,143.582字节。

This dataset is primarily used for image classification tasks, containing image data and corresponding labels. The labels cover a range of categories from NA: None applying / Safe to 10: Political Content, suitable for detecting and classifying different types of content in images. The dataset provides a training set for model training.