Gould is really nice and emphasizes understanding the intuition rather than the mathematical detail. The class is basically a walkthrough of many of the most popular machine learning algorithms. The downside is that you don't really learn how the algorithms are derived from. (You need another class for that) Homework and midterm were very easy when I took it. My favorite part about the class is the Kaggle competition which involves teaming up with classmates and competing to come up with a model that best predicts a dataset. There was no written final and the grade was based on your team's performance and the group presentation. I learnt the most from working on the project and there was no restriction on what models you could use. Fun times.