← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6682.b · 生成模型:GPT-6

分数范数并非满足全部范数公理

理解等级:中级

0时,常用的Lp形式可作某些任务的差异评分,但通常不满足范数的三角不等式。选择这种形式需要明确数学与索引假设,不能因为写作p范数就套用所有度量空间结论。

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

      Fractional distance, fractional norm —p-norm or Lp for 0 < p < 1 City block, Manhattan, or taxicab distance; sum of absolute distance (SAD) —1-norm or L1 Jaccard distance; inverse set similarity Mahalanobis distance Levenshtein or edit distance The variety of ways to calculate distance is a testament to how important it is. In addition to the pairwise distance implementations in scikit-learn, many others are

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第04章《从词数到语义分析》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-4.html) - 结论依据:摘录中关于分数范数并非满足全部范数公理的定义、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:Fractional distance, fractional norm —p-norm or Lp for 0 < p < 1 City block, Manhattan, or taxicab distance; sum of absolute distance (SAD) —1-norm or L1 Jaccard distance; inverse set similarity Mahalanobis distance Levenshtein or edit distance The variety of ways to calculate distance is a testament to how important it is. In addition to the pairwise distance implementations in scikit-learn, many others are