Not enough support from the staff. The assignments are strange, some don't relate to the lectures, some are hard to identify what kind of answer is being expected from us. Discussion forums are a joke, I've managed to go through because of some good souls like Kurt Woshnagg. Great lectures though, content is very nice, but not enough visualizations of the math presented.

good overview of methods but project part was frustrating due to slow Jupyter servers which blocked progress. Overall still positive as course content is unique.

Assignments are poorly designed. Staff is unresponsive. IThe same appened with the first course of the specialization.

Do not attempt this course unless you are extremely experienced in the topic and python already.

the course content is okay. but the coding exam really needs improvement.

This could have been the real deal with so many fascinating topics to learn here, but unfortunately, this specialization is setting new low standards in each assignments. The grader does not work, sometime we are asked to produce wrong results (as oppose to the research material). It is very frustrating!

Good reading assignments.

They need better and more qualified support staff.

Content is good but assignments are buggy.

Content and programming assignments are not much correlated. Lots of kernel problems while submitting assignments and late reply by staff.

Excellent course.

I only wish to have had programming assignment with RNN and Hidden Markov Models instead of three assignments on PCA. Although they highlighted a interesting application in finance.

So far so good. The lecturer refers to projects of which some weren't covered in this course. So a little confusing. Takes lots of googling to finish this course.

Very informative

Completely horrible labs.

And no response on the forums, errors in the labs remains for several months.

This is not acceptable, the course should be removed from Coursera!

It's excellent and incomparable course!

If assignment are clear this course would be a great one. So I would like to suggest that explain more details about assignment and some guide lines

Many technical issues with assignments. Additionally, assignment instructions are often poor or insufficient.

Not Worth the money. Although the assignments is a bit better than in the first course of the specialization, there is no help at all from the coursera team, even when it is impossible to grade the assignment. Do not spend your money there and buy some book instead

content of the lessons is quite good, I would give it 5 stars if the assignments weren't so buggy, contains mistakes, unclear instructions, no help from staff/moderator/instructor, technical issues that are not resolved, etc. a lot of frustration, it just feels like the course was rushed to production and they let the students debug it

The exercise doesn't match the course materials at all.

A great course with deep insight.

This Prof. really have the talent of complicating even the most simplest of the ideas. His teaching method is really bad. Plus some assignments have nothing to do with that week's lectures.

WORST CLASS EVER. Stay away!!!! I want my money back. (and even that is not possible).

the homework autograder does not work. The mentors tell you to simply upload code of which everybody knows that it is incorrect instead of fixing the autograder.

Sometimes those incorrect "fixes" that the mentors give, don't work either. So no way of finishing the class.

Audio of the videos is of very poor quality.

Too many shortcomings and errors assessments. Tests at the end of the videos cut what Igor is saying and they are often about the following video.

The assessments are also very shallow compared to what we are supposed to learn and the 10-people staff is never online and almost never answers any message.

Assignments are atrocious, replete with errors. Staff seems not to care as these have been pointed out and left unfixed for months.

Add some hints in the notebooks, it was very hard to understand some parts

The lectures were truly outstanding, the best overview on different methods in machine learning I have seen so far. The problem sets were also interesting, informative and introduced several useful api from sklearn, tensorflow. With a little work these problem sets could (and probably should) be improved to match the quality of the lectures. For example adding more clarifications in the homework notebooks would be very helpful. Having said this, I think this is an excellent course, and highly recommend it.