知识卡片

普通读书笔记卡 · 7006.b · 生成模型:DeepSeek-V4.1-Flash

特征定义与特征值:两个必须区分的术语

理解等级:中级

作者刻意区分两个极易混淆的术语。特征定义是代码、算法或书面描述,说明要从底层数据中抽取什么信息、如何抽取,它不是某次抽取的具体结果。例如『取客户来源 IP 查地理定位得到国家』是定义,而『俄罗斯』这类具体国家名不是。特征值则是把定义作用到某条数据后得到的具体输出。作者说明这并非标准术语,只为把『怎么抽取』与『已经抽到了什么』分开。这个区分在特征生命周期中尤其关键:定义会被创建、评估、更新与停用,值则被提取、存储与删除。

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

    • Feature and Training Data · 查看原书依据

      Before we go too much further, we need to clarify two completely distinct uses of the term feature and how we differentiate them: Feature definition This is code (or an algorithm or other written description) that describes the information we are extracting from the underlying data. But it is not any specific instance of that data being extracted. For example, “Source country of the current customer obtained by taking the cust

    参考来源

    - 位置:《Reliable Machine Learning》第04章《Feature and Training Data》(源文件:extracted/Reliable Machine Learning/_epub-src/OEBPS/ch04.xhtml) - 结论依据:原文明确阐述该主题,并给出定义、实践或反例。 - 原始内容:Before we go too much further, we need to clarify two completely distinct uses of the term feature and how we differentiate them: Feature definition This is code (or an algorithm or other written description) that describes the information we are extracting from the underlying data. But it is not any specific instance of that data being extracted. For example, “Source country of the current customer obtained by taking the cust