na

Ok

OK

..

Yo

5

1

1

o

a

5

❤

.

I

5

G

t

-

Relatively a good refresher course, although coming from Duke U I expected better. There are too many tweaks (lack of scaffolding) that need to be made from a pedagogical standpoint from weeks 3 and 4. Probably need to make it a 5-week course and not try to cram everything into week 4 just to finish the course. Also, the time limit to take the final test was way too short and needed to be set for at least an hour (problems 2 4 11 12 take at least 20 minutes between setting up probability doing calculations and checks). Also, too many glitches with answers coming up in computer code (Over & Over) which basically buries the learner and forces them to guess if they have no clue what the code means, including the anxiety and disorientation it causes during the final. Many times taking quizzes I felt like the problems were made more for a computer programming course than a refresher course. meant to re-sharpen math skills for data science.

As it is now, the course is a much better resource for reviewing the material (which was fine for me as it was what I was trying to do) than for learning it first time. It would be much better if it had more of the same, which is why I am giving it 4 stars instead of 5. In my opinion, it is too brief; I hope to see a part 2 expanding on the material provided here. Many of the topics mentioned, and they really were mentioned more than really taught, should have been talked about in more detail. I've completed the whole course in about 4-6 hours over 2.5 days. It is a good attempt, but it is hardly a sufficient preparation for the field of Data Science; students looking to take the course should be aware of this.

TLDR: A nice and brief overview of many important concepts (sadly, missing linear algebra) which lay the mathematical foundation for getting into Data Science. Needs to be expanded upon.

First I would like to thank the instructors, Paul Bendich and Daniel Egger for doing such a wonderful job of creating a power-packed course. I really loved the course. I'm a post graduation pursuer in field of data analytics and I was looking to Brush Up my math Skills as you know data science is a multidisciplinary subject with heavy emphasis on math and stat. The course was neatly done right down from technical aspects to content, the video companion sure provided a lot of help. I would like to mention Paul because his enthusiasm and energy was contagious even though it's a video. The only Con I felt was that Bayesian theorem could've been excellent if it had 2 or 3 more different problems that was solved; Granted, that might make the content a little bit longer than a 4-week course but I felt Sets could be little bit trimmed and Bayesian Theorem little more enhanced.

In the end I liked the course. The first part was boringly simple though with a few surprises. The second part was, it seemed to me, absurdly difficult, with the tests requiring a higher degree of skill than implied by the lectures. I think the lectures were theoretically preparation for the test, but in practice were not any where near what would be required for to perform well on the tests. I went to other sources such as Khan academy and the internet to gain the skills I needed to do the tests on the second part. Still, once I had, through other sources, acquired the requisite skills, the lectures made more sense to me and I actually enjoyed them the second time around.

The last section was really tough and a bit more explanation or examples related to the last test would have been appreciated. I have a background of statistics/probabilities knowledge, I imagine students going through that course with no prior knowledge might fail to succeed the tests and give up.

For example: the question about the full house is not intuitive at all and does not seem to be explicated in the videos/video companions. I understand it requires the student to search for himself but this one is really hard, and I had to look for a poker lesson on the internet to understand the calculation of this one.

But thank you for this course, it was great to follow.

The course is intuitive and helps a lot with learning Mathematics (mostly on the statistics and probability side of things) all of which is extremely useful when working on Machine Learning algorithms. But, one major feedback that I would like to give is, the last portion of the lecture with topics:

1 Probability Distribution

2 Bayes Theorem

3 Joint and Conditional Probability

The course should have probably split these topics into separate weeks and elaborated more on these. Just 12 minutes is not enough to understand Bayes Theorem or Joint Probabilities, it becomes difficult and I had to watch many other lectures and tutorials to get a better understanding.