While the course content is very good just like the previous 2 courses of the specialization, quizzes on this course have little language errors that make the questions vague

More coding exercises could be included with much more mathematics background explained. Videos could be made a little shorter. There is redundancy in the some of the videos.

Good overview of how to structure ML projects with great practical advice. I wish the course had includes a programming lab to help us try out and practice some of the ideas

This course is pretty good. In this, I can learn about real world situation on developing machine learning projects. But I think, it could be better if has some code (Maybe)

Good content. The tips and tricks a experienced AI practitioner has was shared. But at least one programing exercise applying all the concepts learnt, would have been great.

In the first quiz, the comments about why an answer is correct are too simple. On the contrary, in the second quiz, the comments are really good and give us better feedback.

I feel more Examples should be given regarding the variable and bais tuning, also Error analysis videos should be a bit in-depth. Everything else is as good as it can get :)

The simulations were very good to build a good intuition about setting up a machine learning project.

But I regret that we didn't have coding exercises. 4 stars for this one.

Me gustó el curso, pero creo que algunas lecciones fueron un poco más largas de lo necesario. Debo agregar que me gustaron mucho los simuladores, creo que me ayudaron mucho

Helps clearly in understanding practical aspects of deep learning. An additional week, highlighting the aspects of productionizing a deep learning project would have helped

Since I myself am working on a few projects, the concepts here are somewhat useful in error reduction. Especially when the models are used to automate medical applications

A unique course to understand the process of establishing a ML project. But lacks tools information and a more structured definition of the process. A bit too theoretical.

Thanks for the practical tips and insights from real projects.

Your pool of heroes of deep learning is very skewed. If the field is so skewed, then it's a bigger problem.

Good course for the ones who already started developing ML systems. This will help us in improving the ML systems and identify what can be done for which kind of problems

Most of the material was quite useful, but some was, perhaps, too obvious. Also, some things were discussed too thoroughly, and, in my opinion, that was a waste of time.

Would be great to obtain more concrete information.

For example, instead of "requires much more training data" to obtain "requires ~1'000'000 samples instead of ~100'000"

Excellent content overall. However, reiterates some of the knowledge already presented in the two previous courses of the specialization. Lacks programming assignments.

I wish there was more examples, visualization and depiction of work with referral to papers or experiment here. or perhaps a bit of project management, data management.

Enjoyable, but felt a little less challenging and more hastily assembled. Regardless, the material is valuable and as always, a pleasure to be instructed by Andrew Ng.

The course material is relevant and useful, however, I agree with other reviewers that these 2 weeks should rather be a 1-2 weeks addition to one of the other courses.

Quick course. Worth taking because gives some practical guidance on what avenues to pursue when finding a optimal model (which takes into account human time required)

The first week is too long with repetitive materials. The second week is very interesting. However, I wish the course was designed such that it required some coding.

Great course, great professor .. the only issue is that I feel sleepy every time I watch the videos :), it's some how single tone. Also the audio could be improved.

Not quite as practical and informative as the first two courses in this specialization, but overall still quite enjoyable and helpful. Excited for the next course.

It is always important to learn above the problem-solving methods and tools. This course teaches the complete diagnosis methodologies for Machine Learning problems