Really good, got stuck on some parts but I got the help I needed, also the last week was very interesting :)

P.d. : The quizzes needs more feedback, because sometimes I didn't know why I was wrong at some questions!

Otherwise an excellent course, but the programming exercises were a little trivial. Then again, wouldn't have a suggestion as to how to make them more challenging without blowing up the required time spent on them.

Great course and content, but some assigments were too "deep" in difficulty. Not that difficult things can't be asked, but the jumps could have been a little better adjusted. But overall the course was very nice.

Lecture material is great. One suggestion from my side would be to spice up the facial recognition programming assignment a bit. To me it felt like all the magic would be hidden in the img_to_encoding function.

Great course content but frankly compared to Andrew's ML and the past 3 courses, the polishing of the notebooks assignments was not there and I wasted a lot of time fixing errors that were not helpful to learn.

The course itself is fine and is probably what most students enrolled for but Coursera is unacceptably broken. Had much greater technical problems with the programming assignments than any other course so far.

I enjoy almost all of the contents in this course. Very nice introduction to CNN. The only problem is that the last assignment for face recognition is filled with bugs. That's why I gave a 4 star instead of 5.

The first 2 weeks are really good and show many important facets of CNNs. I really liked the programming exercises. Week 3 and 4 dive more into certain applications and not all of it might be everyones taste.

4th course in the specialization - the hardest one so far. One minus I think the frameworks' tutorials (tensorflow/keras) are lacking, making it very confusing when you need to do the programming assignments.

Course content itself was excellent, assignments were a bit buggy and I definitely spent more time then I wanted trying to figure out how something wasn't working. Discussion pages are very helpful for that.

Its nice course, it gives you very clear under standing of CNN. I felt exercises could be little better. once again thanks Prof. request coursera to imprive theri suppor regarding questions and other issues.

This was a really nice course with state of the art papers. I enjoyed very much this course. My only suggestion would be to fix the last programming assignment on face recognition and timeouts to jupyter.

Excellent course, I learned a lot! However, I think there is a fair amount of bugs in quizzes, assignments, etc.. Also, the videos require some editing, often some phrases are repeated two or three times.

Great course, but technical difficulties with the grader as well as incorrect solutions and/or instructions led to time being lost misunderstanding concepts and/or performing kludges to make things work.

Though much complex this topic is, it was explained in better manner and architecture of Alex Net, VGG 16 were explained in simplest manner, the box method for object detection is also neatly explained

Assignments in this course lacked in giving intuition. There should had been at least one assignment that makes us comfortable with all the background functions and other things that were kept abstract.

The lectures are PERFECT. couldn't be clearer.

The exercises are very explanatory, but impossible to understand what the grader wants.

I wish the forum's mentors had been more active.

overall great course.

Had much less time to follow this course and I've been less able to focus on convolutional notions, but I still managed to finish the entire course, working on the projects thanks to the clear outlines

Great content in lectures! Automatic graders for programming assignments can be tricky, and set to old versions of tf sometimes, but answers to these issues are readily found in the discussion forums.

This course has been one of the harder ones in the specialization. I found some of the programming exercises to be somewhat challenging. I am really glad I finished it! Time to head to Course 5 now!

This Course was really FuN and hands on.

Although I find out there were general jupyter notebook based problems in the face recognition notebook in week 4. However, the Discussion forms helped a lot.

The programming exercises should be made more extensive. Also, as a personal recommendation, maybe include some of the latest coding models and standards :). Otherwise thanks for an awesome course

The content is not as high quality as the previous courses. Im still fuzzy on some of the material. But this is still world class, please accept my gratitude for this unprecedentedly great content!

Course is good. It will be really awesome if there is a sample example scenario of assignment with code to practice before going to the assignment directly, so that learners can learn very easily.

Great material! I felt the last week was not as interesting or strong as the other ones, but I think I came out with a good idea of what ConvNets are and how to apply them in TensorFlow and Keras.