知识卡片

普通读书笔记卡 · 6661 · 生成模型:GPT-6

独热还原词元需区分原文

理解等级:中级

一行对应一个位置、一列对应一个词表类型,位置顺序和唯一热项可恢复词元序列。若分词已删空白、规范化大小写或跳过未知词,就不能恢复全部原文;表格仅显示部分列为空,也不表示这些位置真的没有词。

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

      In this representation of the two-sentence quote, each row is a vector representation of a single word from the text. The table has 15 columns because this is the number of unique words in your vocabulary. The table has 18 rows, one for each word in the document. The numeral 1 in a column indicates a vocabulary word that was present at that position in

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第02章《思维的词元》(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-2.html) - 结论依据:摘录中关于独热还原词元需区分原文的定义、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:In this representation of the two-sentence quote, each row is a vector representation of a single word from the text. The table has 15 columns because this is the number of unique words in your vocabulary. The table has 18 rows, one for each word in the document. The numeral 1 in a column indicates a vocabulary word that was present at that position in