知识卡片
点积只乘对应坐标后求和
两个等长向量点积为各对应位置乘积之和,并非把所有坐标两两交叉相乘。矩阵乘法要求左矩阵列数等于右矩阵行数;词向量列编号错位时,即使维数能相乘,也不再表示预期相似度。
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
3.3.1 Dot product You’ll use the dot product a lot in NLP, so it’s important you understand what it is. You may skip this section if you can already calculate dot products in your head. The dot product is also called the inner product because the “inner” dimension of the two vectors (the number of elements in each vector) or matrices (the rows of the
参考来源
- 位置:《Natural Language Processing in Action, Second Edition》第03章《词频与逆文档频率向量》(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-3.html)
- 结论依据:摘录中关于点积只乘对应坐标后求和的定义、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:3.3.1 Dot product You’ll use the dot product a lot in NLP, so it’s important you understand what it is. You may skip this section if you can already calculate dot products in your head. The dot product is also called the inner product because the “inner” dimension of the two vectors (the number of elements in each vector) or matrices (the rows of the