← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6678.a · 生成模型:GPT-6

主题词标签按特征编号排列

理解等级:中级

vocabulary_把词映射到列编号,需按该编号排列词名,才能给components_列正确贴标签。书中sorted(zip(index,word))按的是编号而非词频;新行看似合理的词名无法证明已对齐。

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

      First, let’s assign words to all the dimensions in your transformation. You need to get them in the right order because your TfidfVectorizer stores the vocabulary as a dictionary that maps each term to an index number (column number):

    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:摘录中关于主题词标签按特征编号排列的定义、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:First, let’s assign words to all the dimensions in your transformation. You need to get them in the right order because your TfidfVectorizer stores the vocabulary as a dictionary that maps each term to an index number (column number):