This is the practical course.There is some concepts and assignments like: pandas, data-frame, merge and time. The asg 3 and asg4 are difficult but I think that it's very useful and improve my ability.

Overall the course is fine. Much of the work is left out to the user, which would be a good thing if the lectures actually spent time discussing a topic. The instructor picks up a topic and shows us one example and is done with it.

I was disappointed with the teaching style. That being said, I did learn a lot in this course. I learnt a lot of stuff, but I wasn't taught much. Some of the topics were really interesting but they are concluded in 5 minutes max.

I really wish the programming walkthrough were more comprehensive and not just "here's how you do this thing, let's move on".

The assignments are challenging, but are poorly worded. Half the time I had to figure out myself what the assignment was asking me to do.

The assignments took too long for me to complete .

great course, but the auto grader needs updating

Dreadful course. Instructors saw no value in presenting elements of course that would help learners complete the assignments; rather you are sent off to teach yourself about uncovered techniques needed to complete the assignments. From some of the posts from previous students on GitHub, they resorted to deriving the answer from another means (Excel?) and simply providing the answer as a constant value, in order to receive credit for particular questions. Not exactly sterling knowledge transfer, from instructor to student! This course should be presented as a challenge course to people that have already learned Python Pandas from some other venue. (BTW, Pandas documentation is also dreadful, as of this writing.) This is definitely not the way to learn Python for Data Science if you are a busy professional software engineer. (Wish I had a good recommendation as an alternative.)

The only positive aspect of this course is the challenge to work with defined datasets, to complete specific tasks, during week 3. (This was as much time as I could afford to allocate to this course.)

From a 40+ year software engineer, with doctorate in CS, a part-time instructor at a private university, with a very challenging technology job in a multi-national corporation.

Was not a fan of this course at all. The first assignment is completely on regex which I understand that it is an important topic, but that's a fairly advanced topic in data science so to have as the first assignment of the first course in this specialization seems a little ridiculous. Not a single question on the assignment was on numpy which we spent the vast majority of the week learning. I did not get much out of the other assignments either. Dr. Brooks is really not the best teacher. Very knowledgeable, but not good at relaying that knowledge to others in a clear manner. If I could go back a month I wouldn't have taken this course.

It would 3* for materials (they are acceptable but not great until Week 4, when there is only one video crafted by staff and the rest are links to external materials including content for which you have to pay). It definitely would be 1* for assignments. The way they are designed and described (especially in Week 3 and Week 4 ) makes it impossible to solve them without referring to the forum where there are many clarifications. Staff member doesn't see any issue with it. My favorite: asking for a ratio and expecting percentage in the answer.

The assignments are good practice, but the course teaches you nearly nothing. You have to do your own research to figure out how to do them.

There are some very useful Mentors on the forums to help the assignments, and if it were not for them, this course would be unbearably frustrating and useless.

Video lessons go way too fast and don't actually try to teach you anything. If you're already a wiz at using Python to do data analysis, then you could certainly keep up, but then you wouldn't need the course in the first place. Very poorly paced.

This is not an introductory course! There is a very large assumption that you already know a lot of about the pandas library, as well as extensive knowledge about dataframes and series.

the course content felt like a mishmash of basic statistics and paywall-locked articles, leaving me underwhelmed and questioning the value of my investment. It's disheartening to see such fundamental concepts hidden behind a paywall, making it feel more like a money-making scheme than an educational endeavor. The exercises, while intended to reinforce learning, fell short of engaging and challenging. Instead, they were monotonous and repetitive, failing to simulate real-world scenarios effectively. Additionally, the insistence on manually coding names and resorting to external sources for basic information felt outdated and disconnected from modern data science practices. One peculiar aspect of the course was the inclusion of an hour-long lesson on regular expressions (regex). While regex can be a valuable tool in data manipulation, dedicating an entire hour to it seemed excessive, especially considering the limited relevance to the broader scope of data science. However, the most glaring omission was the absence of machine learning and deep learning concepts. In a field evolving as rapidly as data science, neglecting these fundamental topics is a disservice to students seeking to stay abreast of industry trends and advancements. In conclusion, while the course may offer some insights into basic statistics and data manipulation techniques, its outdated approach and lack of comprehensive content make it difficult to recommend. Unless significant updates are made to address these shortcomings, prospective learners would be better off exploring alternative resources for their data science education.

This course is a real waste of time! Please avoid!!

The lecturer in general teaches nothing. He explains some basic concepts. You can learn them in a 5 minutes YouTube video. Then, you should answer the detailed/technical coding assignments. The assignments have nothing to deal with the lectures. The lectures have zero to very limited coding explanation. Then, there is an outdated picky auto grader that grades your work. You will spend hours finding out that your code is correct, but the auto grader works with libraries very old versions. I learned nothing from the lectures but I passed the assignments with 90, thanks to StackOverflow and online resources.

I am wondering who gives this course 5 stars. Fake reviews?

The jupyter notebook made this a horrid experience. Plus Coursera really doesn't want you to bother them with your silly questions, relying on peer-forums. If you scroll through the week's discussion forums, many student posts go ignored.

You can't drop the course past the second (I guess) week so the system will keep on keeping on long after you've given up on trying to figure out the janky notebook thing.

Will not return to Coursera for any reason. Breathtakingly bad experience.

The course lectures hardly covered what was asked in the assignment. For someone who has a full-time job scouting through discussion forums is extremely time consuming.

Too much content condensed into 4 weeks of course. The videos are very fast with ~1.5 hrs every week and do not cover what is being asked in the assignments at all.

Slamming down some notebooks is not teaching. Despite this shortcut does the lecturer still miss to show and explain the difficult cases of the different concepts.

One of the worst course I ever take in coursera. The instructor just writes codes on front of us without explanation.

terrible course please improve teaching efficiency and give a proper realistic assignments

Like many others, I give this course a high rating while lodging a minor complaint that there wasn't much instruction provided. The lectures were excellent, if brief; it's hard to imagine anyone having objections to the instructor. But in terms of teaching the material, it was a bit of a drive-by. The lectures show a few examples, while not explaining the syntax or the various parameters. You have to draw that out of web sites and cheat sheets. If you're not adept at doing that, proceed with caution here. In the end, I was worn out from the effort, but felt that I had gained a lot.

The assignments were challenging for me because this was my first hands-on experience with Python, much less with Pandas. I did not find Stack Overflow as helpful as the instructor suggested. Nor did I find much help in the forums, but that's not quite my style.

My bottom line is that the course was time well-spent, but it could easily have been a six-week course with a more deliberate pace through the various pandas mechanisms such as merging and grouping.

FWIW: My recommendation is to get to know Jupyter Notebook early and follow along with the lectures by opening the Week[x] files in the course download folder. You can pause the lecture while you go play with the code to make sure you understand it. Also, I recommend working with a local version of Jupyter and keep your files local. Otherwise, Jupyter loses connection to the kernel, and stops being able to save your work. The messages are disconcerting, and if you've worked yourself into a frenzy, they can cause panic and confusion. So do all the work on your machine and then upload the whole assignment when you are finished. You upload on the "Create a Submission" screen; it takes only a sec. You won't even have to worry about details like file paths; they'll be the same either way. Once you get the hang of Jupyter, you can settle into a work routine. Learn some of the keyboard shortcuts.

Excellent course. I learned a lot about Phyton, even I thought I already knew what Phyton was, but here Phyton is used intensively.

The tests were really tough. I spent hours trying to figure out how to pass the tests. Also, there is a lot of help in the forums, and a lot of people willing to help.

To be an introductory course I struggled a lot, is a very practical course, and the assignements encourage you to learn more. This is the best technical course I have taken. Lo recomiendo ampliamente

Not nearly enough reference content in lectures. It needs to be made clear students coming from the Python for Everybody course (other Umich course) has a book which I was used to referencing for all of my questions (the class was pretty well self contained and did not require much looking up of concepts). I tried to learn this class the same way I did for the previous one and that totally did not work - I spent wayyyy too much time on my first pandas assignment thinking all of the answers were in lecture/notes. The lecture and notes were very very scant and not well explanative about data structures that are very complicated. Please either write a book or make it more clear how students should learn. Yes, the teacher tells us about stackover flow but I didn't know he was implying for us to use those resources. He should say something like "we don't offer a book with this course so use online resources" and not tip toe around the topic because people paid money to learn so take responsibility and make these changes please. I passed but it was very frustrating at first.

It was informative but i felt the assignment part needed more clarification. I faced the problem that even though my solutions were right the autograder gave me lesser marks. I figured out that we must not print to the console. If that was clearly mentioned life would be easier.

Solid course. I definitely would not recommend it to someone who doesn't have advanced beginner to intermediate python knowledge, though - while it does a good job at a review level for the necessary python, it really moves through the code details quite quickly.

The course is very task oriented so most of the learning comes from the assignments solution, not from the lectures. Succeeding in the course demands a lot of time for the assignments and quite often you would need to google " pandas how to...". If you want a self-contained course, this is not a good option. However if you want a realistic approach to data science, it may be a good choice.