GLMs, CPUs, and GPUs: An introduction to machine learning through logistic regression, Python and…

As a mentee in the ChiPy mentorship program I will be writing a few blog posts about my project -- which was to learn how to implement a couple machine learning algorithms for execution on the graphics card. In this blog post, I'll introduce a few concepts fundamental to machine learning using logistic regression as an example, as well as a code with simple implementation in Python and OpenCL interfaced with PyOpenCL. This post is intended for a broad audience; if you're completely new to machine learning it may be worthwhile to read the beginning and mess around with the code on your own. And if you have any feedback, good or bad, let me know! If you would like to see the code I'm using, you can take a look at my github repository. There are many cases where it feels natural to attempt to model the outcome of an event as the probability that the event occurs, as we often record the outcome of some event as binary data.