Unique awesome course

Thanks coursera

I thought the course was well-taught and interesting, but I felt that it was more of an Introduction -- Here are some things you can do with AI and ML techniques in the context of Medicient -- rather than a detailed explanation of how the techniques work and how to use them in practice, so probably more valuable for Medical professionals than AI/ML specialists. It would be great to have some follow-on courses that get deeper into the technical details; the Coursera Deep Learning Specialization is a great example.

In general, the programming exercises were valuable and engaging, but I have a particular gripe with the grading: In some cases, I had to spent quite a bit of time making micro-adjustments to my program text to satisfy the rather picky criteria of the grader, including one case were I had to remove spaces between tokens in an expression in order to pass. I really think the criterion for grading should be correctness of results rather than conformance of the program text.

First week's programming exercise is really bad. The exercise is very long, but the instructions and purpose of exercises are unclear. For example, I don't understand what the empirical ARR's purpose is, until I finish the whole assignment and review it again. The C for benefit is also extremely confusing, and contradictory. Observed benefit is denoted by -1 while the predicted benefit (ARR) is denoted by positive number. The documentation (e.g. shape of input parameters) is also unclear.