Learner Reviews & Feedback for Applied Plotting, Charting & Data Representation in Python by University of Michigan

GOOD one

useful

heavy

good

good

na

Overall, it is a reasonably good course. Content touches not only how to 'program' a simple / interactive / animated visual but also some theoretic aspects of plotting in general. An interesting thing about this course is that you will decide how challenging the submissions will be though this will not affect your grades. My final assignment for this course: https://github.com/vdyashin/EarthquakesInAsia. In this course, I learned how to create an interactive plot and applied this knowledge in order to create a portfolio-ready visualization.

Though, since this course is about plotting and charting there is a lack of visual materials and great examples of use cases. For potential Russian-speaking listeners, I would recommend sticking with the MIPT-Yandex specialization instead of this one. If that specialization would seem too hard then finish this specialization first. Though, they both specified as an intermediate level. I would claim that this one is for beginners.

First of all, the instructor is very responsible, keep updating information on the forum and course material. The course is a decent level of basic plotting technique review, should be in more detail. Compared with the first course of this specialization, this second course is much less challenging, require less effort to accomplish. The first course is the one attract me of this specialization, the second one, somehow, is a bit disappointing, especially compared with plotting skill of R in another data science specialization, which is even an elementary level course. This course cannot be labeled as intermediate level.

Another problem with this course is the peer review, the grading policy should be changed to punish irresponsible reviewers, no useful feedback got. What kind of responsible one provide feedback in two words, where require to answer three questions (week 4 assignment) to review.

The video guide is pretty good, it shows you a lot of thing that you need to learn. It covers a lot of breadth and depth, but only briefly. For further info, and for the most part of your time when doing assignment, you need to seek the relevant manuals yourself. But that is fine, because matplotlib is very very rich library and there's no way all can be taught in a single course like this, and also it makes you familiar with how to find information yourself.

The main drawback is with the assignments though. I'm okay with the peer review system. The problem is that the assignment specification is not too clear. For example, in assignment 4, you need to think yourself about what you want to visualize. So a lot of time was spent on thinking about WHAT problem to display rather than HOW to address the problem (using plotting/visualization), which is the subject of this course.

This course provides an overview to the matplotlib and seaborn library and guides learners to create useful visualisations with Python. My main issue with the course is that the various topics are not covered in sufficient detail. Successful completion of the assignments required far too much independent learning on commands that were not covered in the course (particularly for Assignment 3).

The course also covered Principles of Information Visualisation in great detail. I thought that was refreshing and useful. However, I felt that the portion on Matplotlib Architecture could be explained in more layman and palatable terms. In addition, it would have been more meaningful if the course drew more on actual real-world datasets instead of histograms generated from a random normal distribution.

I liked the first course in this specialization more. As in the first one, the assignments require you to search StackOverflow, documentations and the discussion forums; videos are nice, but you won't learn a lot from them. Peer review is a double-edged sword. Some reviewers will give quite elaborate feedback, others do not put a lot of effort into their reviews. Peer-reviewing others can be quite annoying as often you have to wait several hours for submissions of other learners. Not to mention the quite large amount of learners who hand in plagiarized code (please look out for these cases if you participate in this course).

This is a whirlwind course that glibly covers some very important concepts without devoting enough time to each one. Week One, although important, should be replaced with more coverage of matplotlib, or a review of the different types of charts and when to use them.

Also, although I appreciate the in-video quizzes, it is difficult to go back and review the concepts you learn from them because they are not in the Jupyter notebook. For instance, there was a method Dr. Brooks used in a solution to an in-video quiz and I could not remember where I had seen it. I stumbled on it again after reviewing the video for something else.

The first course in this series was really good and this one was so-so at best. I got some skills out of it since I obsessively plotted everything and over did the assignments, but the peer grading rubrics are crap. It's all or nothing so if you submit *anything*, you get a grade (and it usually approaches 100%) but I didn't really get much helpful/thoughtful feedback on anything I did since you literally get ZERO feedback from the instructors (nothing!) nor did I get much in the way of helpful info from the people who peer reviewed my work. I find that pretty disappointing really.

The contents of this course are interesting from the point of view of software engineering, but I am not sure if data scientist need such deep knowledge of graphic libraries. The main problem with the course is that the assignments require much more time than the one indicated in the course planning. In addition, assignment descriptions are often confusing, open to interpretation, and lack enough level of detail, which forces the students to begin by investigating what they have been asked to do.

This course introduces the learner to the various design principles that need to be followed while creating effective visualisations that include Alberto Cairo and Edward Tufte's work. It explains the information-visualisation wheel and proceeds to explain how to create visualisations in python using the Matplotlib Library. Various kinds of plots such as Line Charts, Bar Charts, Histograms, Scatter Charts are covered. Seaborn is introduced as additional library.

This course gives an overview of plotting capabilities but I think it could have been presented more methodically. I think the challenge is that there are many ways to generate plots and so this is more a survey of those capabilities. I now know enough to go to stack overflow and matplotlib documentation and figure out what I need to get done, so my goal is accomplished, but my understanding of the plumbing of the different commands feels a big hazy.

The instructions for the second assignment are terrible. My peers graded my assignment based on what they thought the instructions implied I should have done instead of what it explicitly stated so I may have to repeat the assignment and could risk not passing the course which puts my whole specialization at risk. It's ridiculous since I spent sooooo much time on the assignment already due to lack of guidance from the video lectures.

All material, explanations and content are great, no complains there, but I insist with the peer-graded assignments, we don't know if we are being graded well and some people just don't care, take points off for no reason associated with the rubric. Also, I like when the data source is given, I don't have time to search for a source of information that fits my investigation or the imposed topic of the last assignment.

First, I would prefer there is a way to download the sample code professor Brooks used in the course. The screen showing his code moved too fast, and I have to pause and typed to try them out. Second, I will suggest the course show more code examples, more explanation for matplotlib architecture rather than most of the time just verbal description from the profession

The course is okey - lots of fuzzy theory such as Cairo's principles. Interesting stuff, but also quite self explanatory and seemed like a waste of time.

I would say its worth it to do this course if you have not had any exposure to Matplotlib or seaborn, but if you've done any significant using those then this course will feel a bit underwhelming.

Lot of working is required from our side. Not at all an introductory course! More like advanced level course. Requires lot of time, we have to find the documentation ourselves and ask question on stackoverflow. If you are looking for introductory course like Dr. Chucks courses I suggest not to take this course.

This course is really as good or as bad as you make it for yourself, since it is quite bare, more a sort scaffold for you to do your own learning and projects. This is arguably the best way to learn though, which is why I like it. But, it may be challenging for beginners to get the most out of it.

I think length of the tutorial should have been a little bit longer for extensive discussion. The professor should have used words which are most of the cases self-explained or easily understandable from the perspective of audience, so those were somewhat tougher to catch up with professor.

The last week assignment is easier and no strict rules to achieve. The majority will do the minimum to finish the course.

It would be more efficient to push all students to make hard plots in Matplotlib (interactive + animation) with strict rules.

This is not a level of an intermediate course!

The course is great and Mr Brooks is excellent is teaching the concepts. The assignments are fairly complex and they are not very well explained. I would like to see more input provided in the assignments and projects. Also, Iwhat books/websites can be referred for additional readings.