Generally, very good selection of content. The explanations are insufficient for passing the assignments tho, which means that most of the course work is self-study from the web. The buggy auto-grader sometime made the submission of the assignments quite a pain as one has to find a way to change the code in a way that still produces the right answer but doesn't blow up the auto-grader.

It started out great but the more difficult assignments like week3 and week4 could use better wording and better clarification on the expected submission answers. The problems could be more well defined based on expected answers. A lot more of my time was spent troubleshooting the errors of assignment submission than what was being asked to be solved in the assignment question.

Bad :

This course sucks, all the video sound like reading pandas documentation. And I have to google the related explanation all the time.

Good:

All assignments are pretty challenging. It is a great improvement for panda skills after passing these assignments. This is a Enriching-4-week course.

These stars all for the teaching assistant, they helped me a lot on the Forum.

The content and assignments are certainly useful and relevant. However, the lectures are too short and do not help much with doing the assignment. As a beginner, I relied heavily on google and the discussion forum to get through the assignment. And I am not sure if i can actually tackle similar problems again without referring back to the pre-mentioned resources.

Interesting course covering the main introduction topics to Data Science, however there is a too large gap between the theoretical (videos, Jupiter notebook examples, ...) lessons provided and the knowledge required to perform the assignment. The time to do individual research to perform the assignment is tremendous. This is not an easy course at all.

I didn't feel that the lecture material corresponded to the exercises. I spent all my time just looking at Stack Overflow. The exercises are also not clearly written, such that you spend time trying to adhere to solving for the solution as opposed to learning Python fundamentals.

I learned things, yes. But I was here to try and learn what is already there in the books; I thought it would make me understand easily and in interesting ways. I was disappointed. There is no point in taking the course if it sends me back to the books. :(

I have to say this course took me much more time than I expected, because I spent too much time on understanding the programming assignments. Assignment 4 is not friendly to me, because I didn't know about sports in USA.

This is a pretty tuff course and the explanations and teachings are not up to the mark while assignments are good. To complete the course you have to take the help of youtube and web to fully understand the concept.

More concise coding tutorials and less "search on your own on the internet" needed. It is great to get to know the online community and the course needs more coding example directly relat to exams.

A lot of self directed learning, bordering on excessive. Sometimes it takes some investigation to figure out why the autograder did not pass you. Overall, I felt I learned a lot, much on my own.

This is a great overview for python, but the materials/videos/slides are very elementary compared to the sophistication of the homework. Required many more hours than the estimates.

Statistical stuffs are really tough, it's hard to understand without any background also the assignment materials should be discussed more, they should be included in the course.

Recent changes to the course made it a lot worse! Longer, less concise videos, difficult to find course notes, fewer mid-video working problems, and quizzes are pointless.

The videos are ok. The assignments are extremelly centered in data cleansing. The debate in forums is very helpful, and the course staff answers regularly.

Interesting material, poor instruction and not enough practice for things to sink in. Have to basically teach yourself everything through google searches.

I learned a lot by doing assignments, but the course materials are not helpful. Stackflow and Python documents guide me much more than the course itself.

During the course, the instructor was presenting with a background contains moving people which caused a big distraction to me.

The assignments can be better structured. I found it tough to navigate and perform operations

Needs to be packaged in more interesting way..felt course contents and presentation vague

Too fast explanation

Issues:

1. The lectures are a chore to sit through. Dry, slow, and unorganized. The lecture on pivot tables was not used in any assignment. So why have it?

2. Assignments. Ooof. Lets break this down.

a. Autograder is poor. Aside from the oddities that just break it sometimes, the hidden test feedback is lacking. If my answer is off at the 15th decimal point because my dataframe is 226 rows and not 227. I need another assert and feedback telling me that. Instead of a lesson on logic a lot of these problems became frustrating cases of github searching for other peoples passing code and then working backwards

b. Assignments felt rushed. Each assignment had poorly written questions that frequently popped up in the forums asking for clarification. The assignments themselves had odd jumps in difficulty and assumptions. Some would build upon the lectures but other times they would jump and assume that we would figure out the middle. Oftentimes we did, but imperfectly, and the assignments penalized us for that imperfection. For example, if question says clean the data and we do using one of a dozen different ways why are we penalize if we have 224 clean rows, and the answer requires 227. If that level of end accuracy is required, then we need more guidance to achieve exactly that.

3. Forums. Useless. Filled with garbage, and the useful ones are unstructured mess. For one, the autograders output is small grey typwriter text which is undecipherable and the TAs always wanted it posted. This lead to long chains of code blocks and one line responses. I also think the TAs emphasis on posting zero code is wrong. The entire web is built on Stack Overflow, so why not allow code snippets in the forum?

This course serves more as a 'test' for those quite experienced with pandas than an actual educational course. The main issue is that it would be extremely difficult to complete this just with the course materials. You really need to spend a lot of time on stack overflow and with the recommended textbook to make any progress unless you are already quite familiar with pandas. It is then fair to ask - what is the point of the course? What value does it actually add?

The content difficulty isn't the issue its the fact that there is little support from the course in completing the assignments. The content is essentially a whistle stop tour through pandas followed by some tough quizzes and labs with little feedback for when/if you go wrong. I'm really surprised at how high the other ratings are (look at how many go onto the complete the specialization to see a truer reflection!). Again I don't think it is the difficulty per se, there are other courses on this site which cover quite challenging topics (ML, Deep Learning etc...) that do so a lot better than this course. If you have done one to compare against its unlikely you’d give this 5*s.

Instead of doing this course I’d either recommend going directly to the source and read the pandas textbook by the actual author of pandas to get started or do various Kaggle challenges if you want some practical challenges. The only thing this course provides value wise is the certificate.

TLDR - poor design of class makes for a bad experience and wasted time.

Longer - There is a disjunct between what is covered in the videos and what is tested in the assignments, and there are problems with the assignments (the autograder and the extremely dated pandas version supported) . The videos function as a partial reference guide, which is only very loosely what is tested in the assignments. If I didn't pay for the class, I would see value in the assignments themselves. Since I did pay for the class, I expect more value from the videos. For the 3rd and 4th weeks I proceeded more quickly by relying much more heavily on stack overflow and pandas documentation, to the point of sometimes ignoring the videos entirely. As-is the gaps between videos and assignments, on the one hand, and ongoing difficulty of knowing how to navigate the black box which is the autograder (using an outdated version of pandas) are shunted to the forums and other students, volunteers, and assistants. While the forum is helpful (always start there with the assignments, so you will waste less time on poorly designed / assessed questions!), it is poor pedagogy and a bad experience to offload bad course design onto it.

In my opinion not really a course. You get simple hints in the videos what can be done with dataframes and co and 90 % has to be find by yourself in the internet. Very often the teacher says after running a code "as we can see...", but you can see it only if you run the code in parallel, because the teacher does not show the relevant result. For the assignment 3 you have to search a lot in the internet and use trial&error to get the correct result. The notebooks are not very useful for looking up later and you have no slides to download or similar.

The questions in assignment 3 are very sloppy and inaccurate written. Check the forum contributions before starting to avoid a lot of mistakes due to possible misunderstandings.

The speed is very fast (even at 75 %) and all topics are only touched upon briefly. The only advantages compared to Youtube videos are that you have assignments. There should be a lot more exercises in between instead of having after hours of videos only a difficult assignment.