Bewertung und Feedback des Lernenden für Datenanalyse mit Python von IBM

Overall I learned a lot from this course. However there were too many small disconnects in the course, especially between the video voice-over and the slide material. It was like the video presentations were not sufficiently quality checked to make sure her spoken words matched the written words. In weeks 4 & 5 there were a couple of times when she named one function and the slide showed a different one. When one is still low on the learning curve it is very important that the consistency of the material is solid. When there are needless ambiguities in the material, valuable learning time is lost trying to figure out something that is more "container" than "contents". This stalls the learning process and can create a lack of confidence in the material. Good but it could have been better.

The tools are great and the labs are clear. From talking with colleagues it is clear that what I am learning in this course guarantees fundamental abilities for data science entry level jobs. I truly am thankful for counting on IBM for getting the skills I need to participate in the industry of the digital age.

IBM's brand image has a good reputation and inspires a feeling of high-quality, high-impact solutions. It is dissapointing to see the amount of mistakes, typos, and errors present in the labs of this course. It tells me whoever prepared this material - in representation of IBM - was not considerate of the reputation and image they needed to uphold.

While the content is extremely relevant, it offers virtually no theoretical base or context. Those are actually in the Machine Learning With Python course. Reason why I emphatically suggest the staff to change the course order and place this one as the 8th course in the IBM Professional Certification, right after the Machine Learning one. As somebody who is about to finish the whole series, I can say with property that the current order doesn't make sense and, for that, has a negative impact on our (students) understanding, motivation, learning and development. If this course's theory and context were properly provided before, I would give it 5 stars.

Contents seem relevant, and it gives a decent overview of the process covering data wrangling --> prediction models. There's a lot to digest though, and some rationale is not fully explained. Several sections left me with a lot of unanswered questions where I'm not sure what actions are optional in the process, and which are more essential so to say.

However, the labs struggle with technical problems resulting in users not being able to complete, or even restart, them. In addition to this, the labs haven't been proof read which means the text often being inconsistent with the code. This causing unnecessary confusion for learners.

The labs are very good. That is the most redeeming part.

The instruction videos are quite simply, very monotonous and boring - you don't see the instructor and there is no attempt to make the learning stick.

You don't get timely or quality responses in the discussion forums, so sometimes you feel like you are on your own.

The final lab assignment required you to get on to the IBM cloud and set up your account. I get why they are doing it, but it was clunky. You are required to set up on the free option, but the set up is overkill for a relatively small assignment.

Overall, I probably will not do another IBM course.

Solid overview of the applicability and mechanics of various analysis techniques. Video content was thorough and reasonably well rounded.

Labs could use improvement. Lots of technique shown which allows for a monkey see monkey do approach to learning but not much context or explanation of why an individual approach is used or clarification of the intent of the code. For individuals already familiar with the various packages this is probably okay but without that context the take away value of the course is somewhat limited.

I understand the course isn't officially started yet, but it could've been better. There's much to be corrected in the labs as well as the quizzes. The amount of information was a lot, and I'm thankful for the notebooks I have now with steps on doing things, but the material could've been presented in a more cohesive way, this was hard to follow. Also the labs were more intimidating than anticipated (also with many errors). I think this course should be split into two classes instead with more explanation in both.

The course material was helpful, however the labs had several mistakes which I noticed they have been talked about in the forums since long time ago. Also I had strange experience with final assignment grading. At first I failed by a reviewer , I checked my answers and I was sure they were correct, I complained about it and my complaint went nowhere. By resubmitting it again I got full score! I think it would be better to have a more efficient way for grading the assignment accurately.

While the lecture material is well presented and certainly can be followed, the slides are littered with spelling mistakes, and many in important places (code that couldn't run as displayed.) Even the final assignment had formatting issues, and without the discussion forums suggesting removing the confidence interval, it was taking an excessively long time to run. These are generally minor issues that can be ignored, but as a mass, they are embarrassing at best.

The course content was relevant and quite useful. Its the structure of the course that I didnt like. These are the things that could be improved:

QA before sections are finished does not work - one should first go through the section then the mini QA should start

If one is paying for the course, the slides should be made available for download. Its nice to have reference material for afterward because one forgets things. Even more so if you pay to do a course

Many typos, some code does not match text (e.g. text says test sample of 10% but code has test sample of 15%). Where there are questions embedded in the video they often interrupt a sentence which breaks up the flow of the material. Complicated concepts or uses of code are often mentioned very quickly and the related slide disappears from view too quickly.

My peer reviewed assignment was reviewed twice and both times scored incorrectly but in different ways!

Some of the instructions were not clear enough, with a couple of typos here and there. Alot of explanations can be given to the code, e.g. what is for what. Also, before the video quizzes, needs to let learners look at the screen, pause before flashing out the quiz. Overall, good experience. Aside from having some difficulties trying to understand some parts of the module, but able to pick up Data analysis thanks to the course.

So far the other courses in the Data science specialisation contained a final graded assignment. I found them really useful. This course didnt. Also, instead of telling us about all the tools available in the libraries, maybe explaining why we would use them would be better. I could code these functions myself if I understood them, but just using a library seems like it could lead to laziness and a lack of understanding.

+Useful course for beginners. You get to learn basic concepts although these are not enough to get to work on real projects. Another good point is the set of useful libraries and methods presented in the course.

-Downsides of the course are the amount of mistakes found in the labs which are supposed to help understand the theory seen in the videos, but in some occasions can even mislead and mess the students up.

On overall IBM data science professional certificate track: P​ros: Content is just good enough, instructors are good. Cons: IBM watson and the platform given to practise on is awful and has terrible performance and reliability issues, most often doesnt work and had an impact on my test deliverables. I personally overcame those issues to some extent with kaggle's and google colab jupyter notebook environments.

Great coverage of topic, but unfortunately comes with several imprecise (or even planely wrong) explanations in the videos. Video quality (style of presentation) is ok, but sometimes missing things are slightly missaligned or questions show up before the topic/sentence is finished - could use some polishing. The hands-on labs are great though - if the notebooks open or the servers are reachable.

Material to learn data analysis was very good but had quite a few bugs. It was very annoying to review the assignment of a peer because it is not possible to zoom into the screenshot. Furthermore did I need to flag a person because he had copied screenshots and his notebook was empty or only with screenshots but I was still required to review a second person to complete the course.

Not as impressed with this course as the previous courses. My main complaints were:

-Seemed to be some gaps between the lectures and labs

-Some lectures seemed rushed through w/ simple questions, and did not prepare well for the lab

-Pre-written code in labs would produce errors

-Spelling mistakes (i.e. the week 5 "Quizz")

-No final project to conclude and summarize up our learning

The course was not detailed enough. I think the instructor assumed that people taking the course would know a lot about Regression, Correlation and some other statistical functions, that it was hard to understand or follow at times. Maybe consider 'dumbing' down down the statistical functions so that newbies can also follow.

Overall interesting course. Thank you.

First 2 weeks were fine for beginners, but after week 3 where all new different syntax and concepts like seaborn, visualization, Regression models etc etc were thrown in, it got way too advanced for beginners especially when there insufficient and effective practices available to hone the knowledge. Have to spent most of the time self-learning on other websites.

This is an excellent course for beginners in the data analysis and data science fields as it explains deep technical concepts in layman terms along with the Python code for the same. However, not a perfect course for someone wanting to go into conceptual depth or wanting to expand their knowledge of analysis in Python beyond use of standard packages.

Needs to rely less on the cognitive class platform, just host the ipynb files externally as the labs are inaccessible alot of the time. Course only covers regression models, I would've liked to see SVM, KNN and other algorithms. However the course excels in explaining the relevant maths related to regression and regression evaluation

It was educational, but when you pass a quiz, there should be an option to get answers to the questions you got wrong. And the practice exercises were filled with mistakes, particularly week 5. And the instructor was not responding to students' questions for week 5, which was one of the most challenging weeks. That was annoying.

A lot of un addresses subjects. Many mistakes both in the videos and in the labs.

Overall after viewing all the videos again and summarizing for my self everything, I felt a lot better with the material but I think the course is not organized. I also think that it should get into some mathematical subjects more thoroughly.

A lot of errors make the course confusing. Also, the assigments and labs are "too easy"... it is clearly shown in the videos that there is much more to be done, but the course only demands you do about 50% of what is taught. How are we supposed to really learn without practice?? Give us real and demanding projects!