Principles of Machine Learning

I liked these courses and the way they were presented as Steve teaches much of the practical components in R (in my case) or Python and Cynthia provides an enthusiastic explanation of the statistical theory behind this.

Somewhat related to this, I wanted one or two comprehension questions to accompany each video snippet - I've found generally this is the best way to ensure you are getting the main point (usually EdX courses are better about this).

Of all the courses I am doing in the Microsoft data science curriculum these two are courses I might well go back over, just to reinforce the theory behind much of the statistics.

I recommend paying close attention in this course and taking good notes as you will find these helpful in later parts of the curriculum especially the final project at the end.

This course assumes you have a fair amount of stats and calculus knowledge and it's helpful to have a little knowledge of R. It's also good to be technically persistent - I had several issues during setup that were frustrating.

While Dr. Rudin did a fairly good job explaining the math, it sometime became tedious - so I suppose that's what fast forward is for if you don't care about those kind of details.

Because I could complete the exercises, some of the middle sections on tuning was getting too much in the weeds for my tastes.

I was disappointed that the free version have very limited options in testing your knowledge - you could play around with the datasets provided but you were never challenged to do anything with them - often being asked to "answer questions" that you did not have access too.

I jumped to the end because I was interest in general how some of the other algorithms worked.