Taking the Log of My Own i
Log Math Basics for Non-Math Majors
There’s a lot of math I am having to catch back up with and on as I dive deeper into data science, but one of the more rewarding bits of this journey has been in finally understanding the point of taking the log value of a number. I’ve seen the approach used multiple times in statistics courses and in data science programming, but if I’m honest, I never really understood why it was necessary or how it benefitted the process. Today, thanks to the Cornell Data Science Certificate program, I can understand both why working with log values is necessary **and** some of the benefits doing so can provide.
Take a look at the code here, which demonstrates a simple probability via application of the chain rule using a small data set, then manufactures a data set that has enough features that the non-log values get evaluated as zero. I then use the log method to get around this limitation, and finally show that the values I am generating are equivalent to their non-log counterparts.
I’ll paste screenshots below if you just want to see and not execute the code yourself.
