← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9253.a · 生成模型:GPT-6

无监督结果需另找评价依据

理解等级:中级

没有标签时不能直接计算同一任务的分类准确率,可检查结构稳定性、内部指标及任务相关外部证据。不能据此普遍判定无监督准确率比监督学习低;两者可能解决不同问题。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      It is difficult to be confident about the results as no labels are available to compare. Often external evaluation may be needed to justify the results. Unsupervised learning may be parametric or nonparametric. Parametric Unsupervised Learning Parametric unsupervised learning assumes that the population from which the data are extracted follows a certain probability distribution that has a fixed set of parameters. For example, if the

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第07章《无监督学习模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/20_chapter07.xhtml) - 结论依据:摘录中关于无监督结果需另找评价依据的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:It is difficult to be confident about the results as no labels are available to compare. Often external evaluation may be needed to justify the results. Unsupervised learning may be parametric or nonparametric. Parametric Unsupervised Learning Parametric unsupervised learning assumes that the population from which the data are extracted follows a certain probability distribution that has a fixed set of parameters. For example, if the