知识卡片
稀疏表示只存实际出现的项
大词表下每篇文档只用一小部分词,大多数槽计数为0,可用映射保存非零项。Counter计数实际输入对象,不自动分词、去标点或转小写;书中Morgen被展示为morgen、morning多逗号都不符合所写split代码。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
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
This is our first vector space model of a language. Those bins and the numbers they contain for each word are represented as long vectors containing many zeros and a few ones or twos scattered around wherever the word for that bin occurred. All the different ways words could be combined to create these vectors is called a vector space. Relationships between vectors in this
参考来源
- 位置:《Natural Language Processing in Action, Second Edition》第01章《会读写的机器与自然语言处理》(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-1.html)
- 结论依据:摘录中关于稀疏表示只存实际出现的项的定义、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:This is our first vector space model of a language. Those bins and the numbers they contain for each word are represented as long vectors containing many zeros and a few ones or twos scattered around wherever the word for that bin occurred. All the different ways words could be combined to create these vectors is called a vector space. Relationships between vectors in this