I've read almost every article on the web, every question regarding SVM here, but I still don't get how to calculate w and b, how did they appear in formula, what is weight and what is bias:
$$vec{w} cdot vec{x} – b = 0$$
Please explain in simple words/examples/numbers how to calculate b and w
Contents
hide
Best Answer
$wcdot x – b = 0$, equivalently $w cdot x = b$ is the equation of the separating hyperplane.
$w$ is the normal direction of the plane and $b$ is a form of threshold.
Given a data point $w$, if $w cdot x$ is evaluated to to be bigger than $b$, it belongs to a class. If it is evaluated to be less than $b$, then it belongs to another class.