mathaiml5/adversarial-vision-transformers-robustness
收藏Hugging Face2025-12-12 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/mathaiml5/adversarial-vision-transformers-robustness
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含两个配置：detailed_results和summary_metrics。detailed_results配置包含模型、攻击、防御、标签、预测标签和图像等特征，共有3000个训练样本。summary_metrics配置包含模型、攻击、防御、准确率、计数和索引等特征，共有60个训练样本。数据集主要用于存储模型对抗攻击和防御的详细结果和汇总指标。
The dataset includes two configurations: detailed_results and summary_metrics. The detailed_results configuration features model, attack, defense, label, predicted label, and image, with 3000 training examples. The summary_metrics configuration features model, attack, defense, accuracy, count, and index, with 60 training examples. The dataset is primarily used to store detailed results and summary metrics of model adversarial attacks and defenses.
提供机构：
mathaiml5
