I enrolled in this specialization to learn machine learning using GraphLab Create. Half way into the specialization the creators sold Turi, GrapLab's parent company, making it non available to the general public (not even by paying) and then all the knowledge devalued. I wish I had known this and I would have enrolled on a different specialization. The creators still give you the possibility of using numpy, scikit learn and pandas but I had already done a lot with GraphLab create. The time I invested on my nights after work became a waste. I was trying to convince the company I worked for to buy licenses for GraphLab create.

Coursera should not allow folks to create courses that promote a private license course because it would make people waste their time and money if they decide to privatize the software.

Don't take this course, and if you take it then only use GraphLab create when the authors give you no other option.

Teaching style: Carlos was good, Emily is not very clear and loses focus of the topics and often rambles. She seems very knowledgeable but she lacks clarity of exposition when compared to Carlos or Andrew Ng.

The course, and indeed the whole specialization, was advertised as not requiring the Graphlab Create toolkit. This is untrue, as the final programming assignment does require it. The general dependence on SFrame is understandable since it is open source, but requiring any interaction with a licensed product (even if temporary and research licenses are available) greatly negatively impacted my experience in this course.

If you are considering this specialization I would recommend the Andrew Ng course instead and the main reason is that it isn't depend on proprietary ML framework. Despite the good lectures, the assignments don't help you develop the knowledge required for ML developer role.

Taking in consideration the permanent postponing the courses delivery, from summer 2016 to summer 2017, finally the most interesting part of the specialization was cancelled. I'm completely disappointed with the specialization learning expirience.

LDA is bit too much for this course. Either they should have taken a lot of time explaining the things clearly or they shouldn't have touched it. I feel it was not taught properly.

I found this Course less well prepared than the previous 3 modules. Misleading hints in the assignments, code errors, etc... Also, I found the amount of work required higher, which is not in itself a bad thing, just a bit unexpected.

Organized decently, yet tools such as TuriCreate have been associated to a lot of problems with running the assignments. Additionally, it seemed very difficult to receive any sort of assistance if stuck with an assignment or tool.

I took the 4 (formerly 6) courses that comprised this certification, so I'm going to provide the same review for all of them.

This course and the specialization are fantastic. The subject matter is very interesting, at least to me, and the professors are excellent, conveying what could be considered advanced material in a very down-to-Earth way. The tools they provide to examine the material are useful and they stretch you out just far enough.

My only regret/negative is that they were unable to complete the full syllabus promised for this specialization, which included recommender systems and deep learning. I hope they get to do that some day.

I like the course very much. I learnt so many advance concept and real life implementation.. but slightly disappointed by the quiz question please be specific what you wanted us to answer. looking forward for SVM and deep learning material.

excellent material! It would be nice, however, to mention some reading material, books or articles, for those interested in the details and the theories behind the concepts presented in the course.

Great course, all the explanations are so good and well explained in the slides. Programming assignments are pretty challenging, but give really good insight into the algorithms!.

Thanks!

The materials presented are excellent with well prepared skeleton codes for all ML models. Comparing this course to its three preceding ones, this course is more challenging both conceptually and computationally. The slight drawback is that, because of the highly technical nature of the last three weeks' materials, there isn't enough guidance about how one may construct the ML algorithms from scratch, that is, learners with less experience in computing will, more or less, have to accept the sample codes with little confidence about how to (re)write such codes in the first place.

As a result, I believe that learners with more experience in algorithms and data structure (or learners who proceed to learn more about this area) are likely to gain more from this course for at least two reasons: i) they are more comfortable with the complicated ML algorithms; ii) they can improve the algorithms to speed up the estimation time (some advanced techniques are quite computationally expensive, say over 20 minutes).

In general, I have learnt very much from this course and love it.

This course rushed through the material at the end.

Not happy about course 5 & 6 got cancelled.

Again the lecturing style and course content were excellent, allowing us to write fairly complex functions to implement our own algorithms from scratch but also using pre-built functions when necessary to allow us to explore the effects of different variables. The benefits and costs of the different types of clustering were clearly stated. It's a shame that the specialization stops here, as a capstone project with the same quality of these 4 courses would really provide the students with something they can show off to potential employers. The problem most students will have when coming off this specialization is how to implement and deploy your own model into a service like a website.

Another great course and sadly the last of this specialization. I found the material for this course to be the most challenging yet, specifically the LDA module. The programming assignments were all very manageable thanks to graphlab and the very explicit hints provided but I do not feel like I reached the same level of understanding as I did for the previous courses in the specialization. I have grown to enjoy using graphlab and would likely use it going forward if not for the licensing. I am very disappointed that the remaining courses will not be offered and am now in search for another great machine learning resource.

Awesome course. It was great to learn modern tools in machine learning, not just to apply some black-box on data. I also loved the applications that were showed: it is fantastic to see the algorithms in action, knowing how everything works inside. Another exiting ingredient was how the teachers show you the advantages and weaknesses of each method, as well as the suitable places were they can be applied, or even the most popular extensions or alternatives. I was really really great to had spent those months understanding machine learning in this course and during this favoluos entire specialization.

An excellent Course. I was first doubtful about my interest for this Course, having already read mover Clustering. But this Course surprised me: it more than delivered, presented advanced concepts used in real world, always in a clear and engaging approach. The Tutor of this Course is a key component of my appreciation of this Course. To sum up, great content, great materials (Excellement videos, excellent slides, great assignemtns and quizzes - not a single bug!!!) in a very pleasant and engaging presentation. One work... THANK YOU

Very good content, and great practices. Coding a algorithm from the scratch definitely helped my understanding. The more challenging knowledge like LDA and HMM in the last two weeks are not covered well in great details, but I can understand the course design since that the foundation knowledge required to understand of those algorithms are much more advanced than the previous ones.

Overall, I enjoy this course and the specilization overall, except the Graphlab part which is very confusing and rarely used in the industry.

Another super course. Though admittedly (for me at least) very difficult to make within the allotted time given for one period of the Course. Lots of advanced stuff that require substantial studies to really comprehend, i.e., it should never be enough just to hack & run the code (that's the easier challenge). Still have a long washing list of topics coming out of this Course that I need (want) to understand better. But at least the background to do so is neatly provided here. So without further ado ... Applause!

Thank you so much, Emily and Carlos! Really liked all the courses, and I daresay these are the best ML courses available online. Very insightful, and also cover the mathematical part of the algorithms. Since there are now just 4 courses in this ML Specialization, I would mostly jump to Andrew Ng's new Deep Learning Specialization for further studies. But will look out for your remaining courses to be available once more. If and when they come out, it would be great to send out a notification. Thanks!

Pretty good course. Week 4 is somewhat rough going, but well worth going through. I would like to suggest that the instructors spend some time also discussing essential ML tasks such as web scraping, data cleaning, and format conversion (e.g from Wikipedia text to an Sframe format) as from my own experience these tasks consume around 80% of the work of a Data Scientist. Great job, and you get to keep very useful and well commented code with which to implement your own projects!

Like all the other ones, this as well was an amazing course. The topics covered in were the most interesting ones till now for me, as earlier days when I started programming I used often think about problems like these and used to wonder how it was done. Now I feel like I might be able to do them.

Its a shame that you no longer provide the Recommender System course, since that was something I was even more interested in, and its kinda sad that I am not gonna have access to it.

A great course as the other 3 courses in the specialization.This course introduces and make us implement Knn,Kd trees,Gaussian Mixture model and LDA model for clustering and retrieval.The data set is the peoples wiki from the Foundations course and theres a assignment on clustering images too.If you have taken the other 3 an do this with ease and if you haven't taken those i think it will be better to take this course after the other 3.

This course is a very structured and progressive learning. It is an advantage , if we know python . However, one can still manage and explore Machine learning and Deep learning concept of AI. The case study and real life approach keeps your quest on. This is a great initiative and gives us an opportunity to be future ready while at job. Many notion went wrong about AI and the chapters are well designed to keep us engaged while we

Overall, this course covers a lot of materials on clustering methods and algorithms. The assignment instructions are well-prepared. One thing I was struggling with is that I had to install Turicreate on a Linux system, which was a painful experience. It is recommended that the course team advise on how to install Turicreate less painfully on a Windows machine. Other than this, I think it was a great course!