good.

GOOD.

good

Good

It's too bad some people with phds and very poor teaching skills think they can write up some code and feel they are teaching these classes. That being said, it's super cheap and it's very easy to find information online to supplement the lack of adequate descriptions of the topics. Changes that would make me more likely to take another coursera class :

Don't have a bunch of really short videos, combine them into one longer one.

If there is text or code on a slide, make sure that is in the transcription.

Don't have the dumb popup questions that stop the video and make you find the mouse and click to restart the video. Many of us are listening to the video doing something else, I listen over and over. Sometimes, I have to read the transcription to understand what is being said, so I have to stop, get the mouse, click back up to the slides, press SKIP, etc...

If you have an exam, make sure to later send us the answers - e.g. the code that we were expected to write. This is the weakest and most frustrating part of this class. I was not sure how to some things, in part because I wasn't sure what was being asked, to what detail. Even the class discussions showed we weren't sure what data set to use for what. It seems to rely on peer grading, but most of the responses I got from peers was either completely absent or not useful. But thanks for keeping this relatively cheap.

The course covers basic machine learning algorithms and I do not fully understand why it's ranked as an Intermediate one.

Pros:

You will definitely pick up some new skills after completion this course.

There is simple explanation of a basic idea underlying each algorithm the authors present in the course. It's not hidden in tons of math which is good for beginners and is likely to be bad for those who know something about ML and want to get some deeper understanding.

Cons:

You pay for the course and receive a massive advertisement of IBM services. This shouldn't be tolerated.

You are forced to fulfill a final project using IBM Cloud and Watson studio. So, even if you don't want to work with those tools, it's mandatory for you to waste your time on picking up potentially useless skills.

You might encounter with an error occurring during a signing up on IBM Cloud page. In order to resolve the issue you might need to contact their support team, which is quite annoying, time-consuming and, moreover, your subscription may start another month and you won't be able to suspend it if you want to get a certificate.

Think twice before enrolling into this course. I'd better find an alternative.

At first this class seemed easy to follow, but that was deceptive. While I learned some theory (and some mathematics) behind the algorithms we were meant to learn, there was far too little emphasis on how and when to run the actual code. Normally the labs are a helpful part of these courses, wherein I have the opportunity to actually learn code. Not so with this course.

When I reached the final project for this class, I had no clue how to do what we were supposed to do, as essentially, it had not been taught within the course. I had to seek out other sources in order to actually learn the material and make a lot of educated guesses about what I was suppose to do. I suspect (or hope) that much of this will become easier when I re-take Statistics and some other maths (not course requirements), but that won't make up for the deficiencies in the course. Lastly, the typos and other grammatical errors are extremely distracting and misleading (i.e. "lables" -- do they mean "tables" or "labels"? Who can say for sure!).

The laboratory interface falls short of expectations; its quality does not align with the high standard of the content being taught. In my recent completion of the advanced data analytics course, the Jupyter notebook interface stood out for its excellence, maintaining a clear connection with the course material and lab exercises. For instance, the course taught us how to utilize scikit-learn for the train-test split operation, but the exercise content merely presented the code without the context. While the simplicity is appreciated, there is a noticeable disconnection. If the lab exercises followed a structured approach similar to the course content, such as starting with step 1 - importing libraries and modules, then proceeding to data splitting, model building, fitting, and prediction, it would enhance the learning experience. As it stands, this discrepancy may pose challenges for individuals who are new to machine learning as a subject

Like many of the courses, the instructions are not in a format that supports incremental learning and focuses on the mechanics for performing an activity rather than an explanation for why and the reason we are doing these things.

The objectives and measures of success for the final exercise is not clearly articulated, causing me to guess as to what the evaluator had wanted us to do. The instructions said to solve for the four types of methods, but left it to the student as to if they wished to generate graphics, etc. If the only objective was to generate the Jaccard score, F1 score, and LogLoss (as appropriate) to complete the activities, then it should have been stated. In addition, the examples presented in the course labs did not have us generating the F1 and Jaccard scores for many of the models.

From all the courses so far in this certificate this course feels like the least taken care of. Material gets outdated, same typos and bugs according to the forum persist for years, staff replies only very generally. Frustrating issues with online tools they provide when they don't work well for days. I have done most of the labs on the local environment and strongly suggest to everyone else to do the same. Overall I feel like I gained new skills, but it could have been achieved in a better manner. I would not recommend this course to my friends. P.S. And what on Earth is with these forums filled with "Please, review my project"? Any useful threads are drowning among ridiculous requests to do peer review in the course that has automated peer-review system. Jeez.

Even for an introductory course most lessons lacked depth. Usually the broad idea of an algorithm is introduced and then an exercise shows a python call to which applies it. However neither are there any theoretical/mathematical insights why the algorithm works, nor does one obtain relevant practical knowledge. E.g. the course fails to even superficially explain the many options and parameters each algorithm has and which are necessary to actually apply it in practice.

What makes it worse is that there is apparently no support and maintenance for this course: There are tons of smaller and some larger mistakes in the lectures as well as the exercises, however reports of those as well as most other questions in the discussion forums remain unanswered.

This course was definitely informative, but the final assignment grading process was ridiculous. There was way too much ambiguity with the grading criteria. I submitted the same exact assignment twice, the first time I got a 13.5 and the second time I got a 25. This should not be possible. Much like some of the other courses in the IBM Data Science certificate program, I HIGHLY suggest adding an image of what the solution should be instead of leaving it up to people to determine what they think is right or wrong. This turned into an all day process for me and I'm beyond frustrated with the course and relieved I'm done with it.

Very deep with less, almost zero explanations. Not at all for beginners. Either, it has been given as an overview or should completely moved to Professional Segment.

As I remember, at the very first starting of this IBM course series, it was quoted that you need not to know much coding, but what I am observing by end of the modules, it requires lots of coding.

There must be specific guidelines what to learn, what to master before attempting, otherwise it just becomes a mere certificate.

This course teaches people that it is a valid training technique to train on the whole dataset and then use a subset of training data as a testing dataset. I think i would get fired from my job and kicked out of college if i ever said that. Other than that this course is an extremely shallow introduction to machine learning, useful more as a quick preface to what you'll learn in your first semester at college, rather than something that will be of practical use to you.

Pros:

1. Nice introduction to machine learning

2. Videos are not too long

Cons:

1. Code in exercises made no sense

2. In some videos the presenter went into mathematical concepts not needed to understand the technique itself (at an introductory level).

3. Final assignment was difficult to follow as instructions were not clear

4. You will learn more if you read a machine learning book in conjunction with this course

The videos are good, very clear

The lab exercises when compared to rest of the course is not satisfactory because in lab sessions, the algorithms were not explained and lacks Student excercise. It also lacks clarity around when to use which algorithm

almost every lab uses a distinct code compared to other courses i think it needs more commenting i didn' like the final grade either because is very subjective

This course has good aims and covers a lot of ground though explanations given for the materials are often confusing over overly complex for the target audience. There are also spelling errors and formatting errors within some of the course contents, which causes issues - particularly for the final assignment. You'll learn a lot on this course but it's definitely not for beginners.

The whole time taking this course made me wonder why I didn't just quit computer science and pursued art instead; they are going to discuss this course with me and evaluate my progress and if I learned anything, and I just want to cry, dunno but this was really hard to follow and comprehend.

Some things need to be updated. Such and such is "Deprecated" errors, typo errors, and less and less thought put on in designing the exercises. I can feel the preparers dropping the ball towards the end of this course. Current visualization trends should be included like Tableau.

Very superficial. It teaches at a very high level and doesn't go into details in many cases. There are a lot of open questions and I feel as if I just got a taste. It should be the first course in the certificate as an introduction and then there should be dedicated courses.

Some codes were added but no explanation even. They were fairly complicated and should be elaborated on.

Instructions were also unclear, sometimes we have no idea what the assignment wants. I find my self googling the concepts instead of trying to learn it in this course.

Very entry level course. I don't get the point of explaining the math behind the algorithms without forming any kind of geometrical intuition. The math part is also not really good. I also don't get the point, what exactly they wanted to achieve by creating this course.

Good presentation videos is a plus.

However, the total lack of teachers/mentors support, crowd comments instead of properly designed final project is a much bigger minus.

Don't recommend to anyone who really wants to develop skills, not get a useless paper certificate.

There is no support from Teaching Staff in this course, if you have a problem, you will not get any help from the forum. The grading rubric for the final assignment is really vague, which means the peer grades are arbitrary and not clear.

There are not enough exercises on the application of Machine Learning in Python. Also the server containing the exercises is always down. A lot of type-o mistakes.

The only good thing was the final exercise which was really interesting