← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8852.a · 生成模型:GPT-6

独热编码表达类别而非顺序

理解等级:初级

K类单标签任务可将目标表示为K维独热向量:真实类别位置为1,其余为0。这避免用整数大小暗示类别之间的距离或顺序。目标编码与模型输出概率要区分,多标签任务也不能不加修改地套用一个位置为1的限制。

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

      This representation is called a one-hot code, since just one element of the vector is on (“hot”). Each class has a unique one-hot code. We will see why this representation makes sense shortly. The one-hot codes are the targets for the function we are learning. Our goal is to learn a function fθ that output vectors ŷ that match the one-hot codes, thereby correctly classifying

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第09章《学习导论》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c009b.xhtml) - 结论依据:摘录中关于独热编码表达类别而非顺序的定义、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 representation is called a one-hot code, since just one element of the vector is on (“hot”). Each class has a unique one-hot code. We will see why this representation makes sense shortly. The one-hot codes are the targets for the function we are learning. Our goal is to learn a function fθ that output vectors ŷ that match the one-hot codes, thereby correctly classifying