ML Tips and Tricks

I was working with one of MBA Academy students when he gave me a great idea: explore what makes ML models successful. Indeed, the internet is full of articles on how to build this or that model, which is great. You repeat the example, and it works wonders. But then you add your own dataset - and nothing makes sense anymore. Or it makes sense but does not help at all.
So, I am staring this set of articles where I would focus mainly on what to do when the model does not give you any good results. I hope at some stage I can convert this series into a book but this is further down the road. For now my challenge is one blog post a week. 
Starting with a couple of posts on data (pre)processing. 

Comments