Machine Learning From Scratch.

So the AI is taking over. Not really, not yet anyways. However Machine Learning and Artificial Intelligence are pretty much booming. There are dozens of frameworks to choose from and many really simplify entry. However there does come a point where you want to delve deeper and understand what goes on behind the scenes, I did and man, its evil. All that math, all of the time. Anyways I have since started my own open “framework” for creating ML/AI projects. I will admit, it needs some work, but I am at v0.2, so meh. I am however proud to introduce PerleyML.

It can be found on my github, please use and abuse the code as you see fit.

hellzhammer/PerleyML: My personal ML library. (github.com)

There are 2 clustering algorithms, 1 supervised and the other unsupervised (K-Means and Nearest Neighbors). I have also included Logistic regression and linear regression. There is a bag of words class I will expand on in the future. It is solely meant to handle text/string input at this time, albeit, badly. lol. There is a ridiculous number of loops, like they are annoying but the math is checking out so far.

While there is a lot of refactoring and work ahead, I hope this helps someone else understand what it happening inside the black box of AI. Things will change over time, it will slowly take more professional form. This is just a rough draft. I tried to keep things as simple as possible, if a little inconsistent.

Thank you for taking the time to read through and check out PerleyML! I will be back soon with some articles to document how its used currently. Until next time, take care.

*Please note I am no math wizard, forgive any transgression lol*