ML
Feature Scaling
ystc1247
2023. 4. 23. 01:13
Feature Scaling: a preprocessing step in machine learning that involves transforming the input features to a common scale, typically to improve the performance of machine learning algorithms.
It is important to keep features on a similar scale,
ex ) x1 is range 0<x1<2000, x2 is range 0<x2<5;
x1 is divided by 2000, x2 by 5 then both features are on the scale 0<x1/400,x2<1.
This results in a faster conversion process.
** Do not normalize x0 = 1.
** There is no need to do the feature normalization process with the normal equation.
