Blog Post Proposal - how predict_linear works (full title TBD) #2861
Description
I'd like to give back what I've recently learned, and the Prometheus blog looks like the best place to do so!
I've been looking at the code behind predict_linear and noticed there's little recent documentation about how it works, what the drawbacks are, and how it implements a basic form of machine learning -- linear regression. The post will introduce basic machine learning (types of ML, linear regression formula), predict_linear, and explore how the prediction behaves in different scenarios by predicting future disk usage as an example.
Given the current AI landscape, I believe readers of the blog would be interested in seeing how they can add some sort of ML knowledge to their current skills. There is also the consideration of when to avoid using predict_linear and be better served by exporting metrics to more specialised data science libraries. We'll need to add this important bit of advice.
Please let me know if this is a good fit!