← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9378 · 生成模型:GPT-6

点积按对应坐标乘后求和

理解等级:中级

两个同长实向量点积得到标量,可用于加权输入与对齐程度计算。它可为负或大于1,本身不是概率;神经层通常还包括偏置与激活,不能用一个点积就概括全部非线性行为。

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

      C.1.3 Dot product The dot product, also known as the scalar product, is an operation that takes two vectors of the same size and returns a single number. It is computed as the sum of a component-wise multiplication between vectors. This operation is fundamentally a measure of the magnitude and direction alignment of two vectors: >>> a = np.array([1, 3, 5, 7]) >>> b =

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》附录C《附录C:向量与线性代数》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/appendix-c.html) - 结论依据:摘录中关于点积按对应坐标乘后求和的定义、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:C.1.3 Dot product The dot product, also known as the scalar product, is an operation that takes two vectors of the same size and returns a single number. It is computed as the sum of a component-wise multiplication between vectors. This operation is fundamentally a measure of the magnitude and direction alignment of two vectors: >>> a = np.array([1, 3, 5, 7]) >>> b =