Learner Reviews & Feedback for Machine Learning: Classification by University of Washington

Last portion was a little difficult to relate to why we started this move for large datasets in the first place. I had to keep going to the fact that I am going to be handling large datasets. Like the use cases. simple and effective. The quizzes were simple and the graph questions were really helpful in gauging my understanding of math behind these models.

The contents are very interesting and well explained. Nevertheless, unlike the Regression module, the current one suffers of some technical problem, like slides not well formatted, noisy audio in some video, weekly work load not perfectly calibrated. Despite all this, if you are interested in the subject, you will definitely love this course!!!

I loved the course. Carlos Guestrin is an excellent and engaging professor that really captivate me to work hard to accomplish the assignments.

I just suggest that the assignments should be divided into small pieces to be taken as long the week is accomplished. I felt bad some weeks that had a lot of videos to watch before the first assignment.

Good, funny and super-clear professors introduce you to the main classification techniques out there (except for neural networks). Great if you are approaching this field and want to know more before deciding if you really want to invest a lot in it. 4 stars because it would have been better with more popular python packages than GraphLab.

Not as evenly paced as the first two courses. Also some material was covered at a very high level, whilst I found that some explanations did not immediately build on my understanding gained through the foundation course, but rather confused it. Still a worthwhile course nonetheless. I look forward to the rest in the specialisation.

The course is perfect for people who want to gain in-depth knowledge of classification algorithms but exercise descriptions are vague. I found trouble understanding the flow of assignments. Also, Bagging and Gradient Boosting techniques were not covered under ensembles. Overall, the course is awesome.

The course 3 got pretty technical pretty soon. Enjoyed the first 2 courses without feeling overwhelmed. But course 3 was challenging. I suppose building the expectation of what is to come can reduce the challenge and lead to faster and more number of course completions.

Just one comment about how the course could be improved: the assignments should be more hands-on with fewer pieces of code written in advance. I say this is even though I am not a skilled programmer. The assignments would be a bit harder, but also a bit more rewarding.

Sometimes is difficult to understand the concept behind Classification because some videos are more practical than theorical, I mean it could be better to start the video explaining some concepts and then show and explan some samples and theorical issues.

Thanks.

The course itself is well structured and introduce gradually the complexity. Unfortunately, the exercises requires the use of a specific library, instead of scikit-learn and numpy. Furthermore, they also required Python 2, while Python 3 is now widely used.

As with all the courses in this specialization: great production values, excellent tuition. Useful assignments, even though the reliance of Graphlab Create is a bit of a drag. I also would have liked to see some discussion of Support Vector Machines.

This course covers all the core algorithms used in Classification models. If you have a basic understanding of machine learning, this course can help you build your understanding of classification on a deeper level.

Using discontinued Graphlab in the programming assignment is a minus and low activities in the forum makes hard to find assistance from the communities or mentors but the course material itself is just great.

Very good content, very well explained... great course. Classification its a very broad topic but i think this is great introduction.

The hands on where kinda on the easy side... but very interesting.

Not as good as the previous courses in this specialization - I agree with those who have noted that this one seemed a little rushed. However, these are still the best courses I've found on Coursera.

Great knowledge about machine learning fundamentals, More math illustration needed though it's great knowledge and very great basics about different machine learning algorithm used in reality

Extremely clear and informative. Good introduction to ML. I felt the labs could have had us write a little more of our own code, and would have been better to use non-proprietary libraries.

Presented content is rather clear and instructors are rather easy to follow. Only the assignments are often confusing as there are questions which refer to missing content.

Sometimes the assignment description was hard to follow along. Overall, the course equips me a good understand and practical skills to tackle classification tasks.

I learnt a lot, but I feel course 2 was very well made and this one felt a bit unstructured in comparison. Also, assignments in this course were made very easy.

Assignments were a little too easy, considering that students are expected to have taken the first two courses in the specialization. Otherwise, great course!

Good course overall. Some difficult materials such as boosting were not clear enough and I had to look into a few online resources to really understand it.

Very nice lecture & materials. The only slight negative component this lecture contains is the library used for the programming assignment.

This course could be better if more programming content was taught. The programming assignments are difficult and not taught in courses.

Great course. I'm really looking forward to learn more about clustering in the next course since I know nearly nothing about clustering.