Introduction to Machine Learning

I obtained an interest in Machine Learning when I first learned about it back in my post-grad studies from a random fellow I met at the local campus pub. When I later looked into the topic myself, I came across the now popular MarI/O video on YouTube and then for the next 7-8 years as even more sunk-cost fallacy grew within my career path, I couldn’t stop thinking about it. So when I was ready to officially dive in, I was pretty darn excited. Once again, however, I found myself having trouble knowing where to officially start on the topic as a self-learner. Eventually, after far too many YouTube videos, I decided to “just start doing something” because I concluded that it doesn’t matter if I don’t choose the most optimal starting point, what matters is just starting somewhere at all. Therefore, since I enjoyed my dive back into Python and thought the course was decent, I figured another course from the same fellow would be a good start:

And I was right! It was a solid introduction to Machine Learning, but even more importantly it was a great introduction to some of the tools used in the field. Just learning Numpy, Pandas, and Seaborn was enough for me to be satisfied with this course. Unlike the first Python course I took this one didn’t have milestone projects and instead had an assignment in a Jupyter Notebook after each topic which was a good way to get a feel for the various algorithms and reiterate that you always need to prepare your data correctly for what you’re wanting to achieve. However, I found it wasn’t a good way to actually learn to build machine learning projects and was more of a “this tool exists -> here’s an example of it -> do assignment that copies the example” which meant there wasn’t enough critical thinking for my liking. I instead found it to be more of an introduction class than a master class, but perhaps that’s just me.

I did enjoy learning about all the tools in general and went down several rabbit holes because of how interesting some of them were during this course which I’ve since organized a bit into a GitHub Repo. Ultimately, I did not do a “final machine learning project” at the end of this course because I didn’t feel like I learned enough to make anything of substance. All I’d have probably done is apply different algorithms on some data I’d have to dig up and I want to do more than that. So… would I recommend this course? I would only if you consider it an introduction to machine learning tools in Python. I would not if the expectation was to be fully capable of building machine learning projects by the end of it. Therefore, I decided I’m going to do the popular course by Andrew Ng and hope that by the end of it, I’ll be ready to build a project. I’ll report back on how I find this course later!

Leave a Comment