As I worked through the Module 1 lessons and Module 1 project, something struck me as interesting: the recommended technique for transforming features to make them more normally distributed. The lessons say to apply a logarithm function and show us how to do so. Since it’s been a while since I had been exposed to the math, I thought I’d take a deeper dive and see how log normalization works behind the scene.
While working on my Module 1 project, the King County Linear Regression project, I wrote some functions to automate and simplify certain tasks. I thought these might be useful to me in the future, and maybe even to other people. In this post I’ll walk through some of them.
Think of all the downtime in your daily life that goes to waste. That hour on the subway, that afternoon at the dmv, that 20 minutes at work when the servers are down and you’re waiting for the IT department to get things up and running again. Would’nt it be great if you could use that time to get some data science work in?
As I was working through Module 1, Section 6 of the Data Science track (i.e. the section on visualization), I was struck my something I found peculiar.
I am, among other things, a lawyer.