# What is VC dimension of linear classifier? How to measure it?

I came to know about Vapnik–Chervonenkis during my study on ML. All I get it to represent the power of a classifier. But I don't understand how to calculate it exactly.

Given N set of data points, by using a linear classifier to classify this data points. What is the VC dimension for this classifier? From My understanding, it may be 2^N! But not sure.

I was actually reading a blog post on this topic and they showed some co-relation with the posterior distribution. What is the posterior distribution for a given set of data point N that if a model has the following prior,

x1,x2,…,xn∼N(μ,σ20) and μ∼N(m0,s0)


where m, s0, and σ20 are known.

Is this a,

• Beta Distribution or
• Normal Distribution or
• Gamma Distribution or
• Inverse Gamma Distribution

Any help will be highly appreciated.

## migrated from stackoverflow.comApr 6 at 13:03

This question came from our site for professional and enthusiast programmers.