Overall a good course; I learned a lot. But hard going at times for someone new to Python and Jupiter Notebooks. The time estimates for the module assessments are way under (maybe reasonable if you are already a Python expert and have some familiarity with the relevant libraries, but that's not my situation). File location mismatch between Assignment notebooks environment and submission / assessment environment was very frustrating.

Principles are good and well presented, and the excercises well made and challenging. Unfortuantely the Python is getting way of date, making it difficult to use the answers you may find when looking at the docs of scikit-learn, pandas etc, since they create incompatibilities with the ancient dependencies expected by the autograder.

The content just needs a modest amount of love to bring it up to date with current tooling.

Since there are many theoretical concepts in this course, like model evaluation and tuning parameters, it would be much better if those are explained using real or semi-real life problem examples. Especially the quizzes needed more context as to why a particular situatrion might occur, and why that particular variable of interest is necessary.

Course was good but It became very tough, because me and other students are facing a lot of problems and errors in assignment submission. Submission of assignement is not clean as it should be. please update and remove errors we are facing during the submission of the the assignment.

Good overview of methods. A bit too intense at times though, may have been better to really focus on a couple of key concepts.

어려웠어요 ㅠㅠ

First two courses of specialization were so good, but I am disappointed by this one i.e. Machine leaning. I know this course is applied but then also advice for others, this is absolutely not for beginners, because there is too much rush in this one. I didn't understand 60% of things because new concepts are taught one after another without deep understanding and mathematical concepts that how it is working.

So many technical issues are there. I was not able to move further in this course on time my motive to take this course was related to my academics but now my academics ends but still this course has so many issues in assignment completion.

It was too easy

A great introduction to the practical side of machine learning, particularly if you have already taken Andrew Ng's course. It covers a *lot* of material and the pacing is *very* fast. Week 2 is particularly long, and if you are still a student/working it may take an extra week to complete the course. Quizzes and assignments are not terribly difficult, but be careful of the project assignment in Week 4 (though the bar for a 100% is quite low!). Finally, the accompanying Jupyter Notebooks are very helpful and there are many helpful links to outside resources as well.

A few of the lecture videos feel like an early draft rather than production-quality, with lots of time spent on repeating phrases. The instructor mentions things to be covered "later," but that "later" never comes (for example, in discussing Grid Search). For some background, this course appears to have been repeatedly delayed before its release. To me, is understandable that the creators wanted to get this course out given the demand, but the rush is felt.

Ultimately, however, this is still an excellent introduction to Python Machine Learning, and I do feel the course is well worth taking. Just be prepared to do some more individual learning; however, shouldn't one always be for an online class?)

TLDR : This is truly an EXCELLENT course if you already have a good theoretical basis in machine learning and good skills in python programming. Otherwise this will not be a pleasant experience for you.

As far as I am concerned, I worked it along with several books and this course helped me learn quick and effective hands-on machine learning skills, to complement my theoretical knowledge. If you are a complete beginner in ML, this is clearly not a stand-alone course and you will need, for example, to refer to either Andrew Ng's coursera course on machine learning or to Christopher Bishop's book (as I did). Overall, I consider this course has helped me a lot and I learned a huge amount of useful things and good practices. I now feel confident enough to apply for jobs in the ML field, which is what I enrolled (and paid) for. Nevertheless, I would have appreciated a dedicated section specifically on how to handle categorical variables. This matter is not really treated throughout the 4 weeks and I think it would have been a better choice to include it instead of the very superficial optional introduction to Deep learning. Anyway, thanks for putting up this quality course, it was a very good experience to me.

Muy agradecido, mis felicitaciones al Profesor Collins-Thompson, se muestra como una persona amable, dinámica y con alto grado de conocimiento, gracias a sus enseñanzas estoy aprendiendo más sobre el proceso de machine learning, siento que aun me falta mucho por recorrer, sin embargo, a lo largo de este curso aprendí los métodos, tipos de modelos, herramientas tanto para clasificación como regresión enfocándome en el área. De igual forma la literatura es muy interesante, se encuentran artículos que al leerlos vas comprendiendo como ha sido el proceso de transformación en este campo y gracias a esto, se me han ocurrido ideas que me gustaría compartir o estructurar para evidenciarlas de manera mas formal.

Muchas gracias por el apoyo, gracias por las observaciones y anotaciones dentro de los foros de discusión, siento que puedo seguir aprendiendo mas y es por eso que estoy agradecido por mis conocimientos adquiridos, los cuales siempre puedo retroalimentar viendo el curso nuevamente cada vez que lo considere pertinente.

5 starts for content. The lecturer and slides were good. The assignments were often difficult and took many hours longer than the stated 3-4 hours. Assignment 4 was particularly heavy in time. I finished the course feeling equipped and confident enough to take on straightforward machine learning projects from start to finish. I've dropped a star because the autograder uses an older version of Python and older libraries, which meant I had to spend around 8 hours re-engineering my *correct* code to conform to old libraries.

Addendum: I've uprated the course to 5 stars after having just completed the fifth, optional week on unsupervised learning. It's unassessed but does give a nice introduction to the subject. Thanks!

Though I would have liked a bit more insight into the actual algorithms behind machine learning, this class did a great job of giving us problems and forcing us to be resourceful and hunt down the answers, whether via course forums, Stack Overflow or other random Googling. We were exposed to a ton of different algorithms and libraries, and we got to experience the whole spectrum of data science: data importing, cleaning, exploratory analysis, feature selection, model selection, parameter tweaking and even some visualization. It was a lot of fun: challenging at times, but oh so rewarding in the end!

Very nice class for people who have some intermediate knowledge in Python and who want to dig in, or consolidate their knowledge in Machine Learning. Great overview over scikit-learn, also going into details, and I also appreciated the part of the class about model evaluation. First week might seem not overly difficult, but the intensity of the class ramps up significantly in week 2. For me the level was challenging enough, without being overwhelming. I enjoyed taking this class and obtaining my certification at the end was a very nice reward. A big thank you to University of Michigan.

This course is excellent. It contains a great deal of instruction each week (1-2 hours), and it also has many supplemental references for people who want to go deeper. The quizzes are actually very challenging, and require study of the material. The assignments were easier for me than the other courses in this specialization, but they were focused on application of the material to real world problem, which is the purpose of the course. The final assignment is very instructive and challenging. The instructor is very knowledgeable, and teaches in a thorough, but easy to follow, manner.

The course was great. The final assignment was especially useful as it was almost completely unguided and gave us a dataset which is unlike the tidied up, dummy datasets you'd find in online courses. So it was, by some means, an independent project. The data was messy, full of errors, and maybe downright ugly. We needed to clean the data and do quite a bit of preprocessing to get it in a shape suitable for fitting a machine learning model. The project gave a taste of how a real-world machine learning project might be taken on. Thank you very much, Professor Collins-Thompson!

I completed the Machine Learning Specialization Certificate before taking this course. This course is an excellent applied course that quickly gets into the key aspects of using sklearn. This course is ideal for both new learners and experienced learners who just want to learn more/revise about machine learning. For the final assignment, it requires substantial data cleaning techniques covered in Course 1 in this specialisation. Overall, I feel very comfortable with using Python for any reasonable size of machine learning problems after taking this course.

The class provides a perfect introduction to the scikit-learn Python module. The videos are engaging and insightful. The quizzes are challenging while not requiring too much time writing out solutions (it does take time finding some of the more subtle answers, by reviewing details in the videos). The executable modules are perfect to bootstrap machine learning projects. Homework assignments can get complicated, and you should be familiar with advanced data structure manipulation in pandas and numpy to make progress. Assignment grading is very well done.

Great course on the basics of machine learning. I'd say this course is a great dive into sklearn, which is actually great for many purposes. It barely covers Neural networks, which are the hot topic right now, but it gives you a lot of tools that will suffice in the vast majority of cases, and teaches fundamentals that are also applied to deep learning if one decides to go forward and learn other libraries like tensorflow. All in all, a great addition to anyone's toolbelt, be it engineers, scientists or people trying to jump to a data science career.

Excellent course for Machine Leaning. Discusses wide range of Supervised machine learning and gives a very brief introduction on Clustering algorithms(Unsupervised). Users can immediately put to use the knowledge gained during the course.

Some more briefing about feature transformation and other such elements can be included in the course material to make it better. Also unsupervised machine learning could have been included with grater depth. Overall this course is highly recommended to aspirants interested in ML with some python knowledge.

The course was very well prepared and the instructor presented the material clearly and informatively. I've seen some courses where you spend more time trying to understand and keep up with the instructor. In this instance, this was not the case and you could spend more time understanding the material. The instructor spoke slowly and clearly.

I do have to say I purchased the corresponding book as recommended but I didn't feel it was necessary. Good book, I just think the material in the course was presented well enough on its own.

The course provides a good overview of ML techniques and potential gotchas, and then goes into a real life example which helps round up the theoretical overview with application to real world data and their challenges. This provides a great introduction to ML which positions you to delve into it in much more detail and help in your journey as a Data Science practitioner. Must commend University of Michigan on coming up with the fine balance of theory and practice, which is essential in this rapidly changing space.

Great course for Machine Learning Algos. This series of lectures also helped me in understanding two beginners books for ML -

1. Introduction to Machine Learning

2. Hands on to Machine Learning.

Professor taught in a very informative and easy to understand way. Really thankful to the professor. Each and every algo is well explained with strengths, weaknesses.

questions in Quiz are very good these were not so easy and not so tough.

I will recommend this course if you want to learn ML using Python.

Thanks a lot, sir.

It is such an interesting and practical course for machine learning. If you are looking for courses which allow you to apply what have you learned in practical problems, this is a very good option to consider. I liked how this course is structured, it teaches you the theory first, and then ask you to use what you have just learned (of course, not 100% coverage), which definitely provides a valuable learning experience. Highly recommended for someone who is interested in data science in general.