知识卡片

普通读书笔记卡 · 9247 · 生成模型:GPT-6

多项式回归可对系数保持线性

理解等级:中级

二次模型ŷ=b0+b1x+b2x²是输入曲线,却仍对系数线性,可作为构造特征后的线性回归。原6.10漏掉b1后的x,不能照抄成两个常数项;曲线拟合好不保证远处外推合理。

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

      The yield of a chemical process at different temperatures is shown in the Data File: YIELD1.MTW. The fitted line plot of the temperature and yield in Figure 6.5 indicates a nonlinear relationship. The plot shows that the data can be well approximated by a quadratic model.

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第06章《监督学习模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/19_chapter06.xhtml) - 结论依据:摘录中关于多项式回归可对系数保持线性的定义、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:The yield of a chemical process at different temperatures is shown in the Data File: YIELD1.MTW. The fitted line plot of the temperature and yield in Figure 6.5 indicates a nonlinear relationship. The plot shows that the data can be well approximated by a quadratic model.