fengdu78 / Coursera-ML-AndrewNg-Notes

吴恩达老师的机器学习课程个人笔记
31.55k stars 10.57k forks source link

week2.md #17

Open czstudio opened 6 years ago

czstudio commented 6 years ago

4.3 梯度下降法实践1-特征缩放 最简单的方法是令:${{x}{n}}=\frac{{{x}{n}}-{{\mu}{n}}}{{{s}{n}}}$ 应为 ${{x}{n}}=\frac{{{x}{n}}-{{\mu}{n}}}{{{s}{n}}}$,其中 ${\mu{n}}$是平均值,${s{n}}$是标准差或范围。

hustlong commented 5 years ago

视频里是:s1 is range of value. and by range I mean lets say the maximum value minus minimum value. 所以应该就是该特征的取值范围。

hustlong commented 5 years ago

或者是 standard deviation of a variable. [囧]