The sixth week of Machine Learning was very abstract and subjective. I’ve been seeing things getting more and more subjective as the weeks progress. This makes gaining deeper insights harder and leads to subject mastery being based more on experience.
During the course of this week, we learned how to evaluate the data set that we have on our hands, and to step back and take a look at how our hypothesis performs before we begin using our hypothesis. With a handful of metrics to test out the robustness of our hypothesis and check if we’re over-fitting or under-fitting our data, this week’s material provided the tools needed to improve the overall performance of our Machine Learning algorithms.