Prof. Ng is a very good teacher and the course is content-rich and well organized, but there are two things that could be improved. First, there are many hesitations and reformulations that could be removed from the videos. Second, and most importantly, the assignments are a bit too easy. The answers are almost given in the questions, and if you know how to translate equations in Python you can manage to pass without really understanding what you're doing.

Excellent course, but some of the most technically difficult material I've ever encountered. Gave me a solid understanding of the theory behind CNNs and their applications. However, it doesn't go deep into how to program using tensorflow and keras and holds your hand through most of the assignments. Overall I think you'd have to supplement this course with some keras tutorials and lots of practice to be able to implement any of the assignments in the wild

It was an amazing course that explained everything very well, I am fully satisfied with the learning experience and quizes. However, the only thing I wish to talk about is the coding exercises, at times they feel too assisted while at other times it feels as if we are left not with a lot of information to work on. While the former created a dependency on these instructions, it made it very difficult to progress sometimes when the latter was experienced.

I think I learnt the most from this course compared to the other deep learning courses. The material was well presented and the labs were also very hands-on and had enough to help understand how to implement real world problems. There was an issue with the last assignment where the code which matched the output failed the grader (triple_loss) which I think should be looked into.

Many thanks Andrew Ng and his team for another very well organised course.

Great intro to CNNs and their applications. If you're interested in anything regarding image processing or computer vision, this is a must!

Only negative I have to give is the same that applies to the other courses, which is the TensorFlow part is taught without much explanation and you either copy/memorize the code or you bash your head trying to understand how it works. This course also works on Keras though which I felt was taught so much better!

excellent course, great lectures and interesting excercises - the only dowside is the non-existent difficulty of the quizes / prog assignments, you should pump it up a bit (if the students are forced to think through the algorithms they better understand them, in the current state everything is prepared, you just have to fill in some tiny details - you don't even have to understand what you are doing to net a full score, which is really sad)

The course is really of amazing quality. The only reason I am giving only 4/5 is because there several slide pack not available for some video, and coursera/instructor could have made it easier for the student to follow along with a proper slide pack and taking note. Instead we have download 2-3 slides per video, concatenante them all, or even take screenshots when slides are note provided. Appart from that bémol, that course is gold.

Great explanatory course about the idea of convolutions.

Theory is extremely fine as always! Esp nice to hear about one-shot learning technique and triplet-loss "family"

For the practical things, i'd like to say that it was ~ 3/5. Valuable example would be an assignment of week 4 about making a neural style transfer. Although i passed all 4 graded functions, i ended up with non-working neural net. I mean a lot was uncovered with grader

I really enjoyed learning and certainly appreciate the effort that went into this. The only thing that I would change would be the addition of exercises to help reinforce the TensorFlow/Keras programming pieces. For example, I found it confusing to understand the execution of some of the more complex graphs. I do believer that deeplearning.ai has a new series out focusing on TensorFlow implementations, so this may not be an issue.

It was superbly full of information i was not privy to before now. Convolution as an operation and it's uses are now obviously apparent to me. It could do with a bit more transparency in the code as sometimes I would personally like to experiment on my own but helper functions which i used in the course are not immediately available to me. My most helpful course so far, it gave me the confidence to attempt my first kaggle competition

The course is a great introduction to convolutional neural networks and makes the subject tractable. At the same time, it is in no way a "deep dive". The assignments could be a little better, requiring more from the student. Also, the videos are not edited, so I often heard Andrew's errors while recording when I was watching. The non-editing part is why I cannot give this course a 5-star review. But rest assured it is a great course

From the theoretical point of view it is a very instructive course. What did not convince me very much is the way in which the programming exercises are proposed: there are some passages that are really hard to understand for a person who has never used TensorFlow (like me, even if I passed all the previous courses of deeplearning.ai) so I would prefer some more hints on the sintax and how to use certain functions of this framework.

Very in-depth explanation of how Convolutional Neural Networks work. You will pretty much learn all the theory behind them and all the theory behind several systems like face recognition and neural style transfer. You will even get to implement an object detector! Although the object detector part is mostly done behind the scenes but it still teaches you the basic building blocks of the state-of-the-art algorithms in this field.

An exceptional course with a great deal of useful architectures and design ideas. Before this course, I had no idea what residual and inception networks are, however, the course gave me a relatively-deep look inside these networks. 1X1 convolution (network in network), convolutional implementation of sliding window and lots more are used efficiently. Face recognition and verification is one of the lovely topics that was covered.

I enjoy the content and assignments. Andrew is a great teacher. My one complaint is with the assignment notebooks. I find them very glitchy. On the final happy house assignment, I think I spent more time trying to load and reload the notebook (when I get the "method not allowed" warning) than actually finishing the assignment. And I often had to copy my solutions into a text editor so that I would't lose them on the reloads.

A very good course overall and i learnt lots.. But I felt there were too many details to be covered and hence lots of it was not presented in enough depth.. this would lead to a bit of confusion at times- at some places, i would find myself getting a bit lost on how this happened or where this came from., or what the outcome of such operation/convolution would look like in high dimensions or a different scenario..etc...

I generally liked the course very well. However, one could tell that this was the first time the course was given as there were a couple of bugs in the programming-assignments. These were often not easy to understand, not least because the grader-feedback was usually not very helpful. I expect that these bugs are fixed in the next iteration of the class in which case I would give 5 stars. As always very insightful and en

Great teacher and material. Sadly it seems like that team is rushing the later parts of the course and it has quite a few errors and issues. The issues in the programming assignments are specially aggravating since they are intended to validate the knowledge acquired and it's frustrating to have to resort to trial and error to fiend the solution that the grader likes and not the solution that is appropriate or correct.

Everything else is perfect, except the W3 assignment submission took me 3 extra hours to be correctly graded (my results match the expected results, and I didn't violate any rule that FAQ mentioned, but grader just gives 0. I read through the forum threads, tried all approaches, and finally get it passed after having Coursera helpdesk load a fresh notebook for me and type in all over again, a little bit frustrating).

Very interesting and well structured course. Great lectures and content.

Only critics: The actual tasks to be computed/coded in the assignments did not include the "interesting"-ML issues. A lot of linear algebra coding was asked for (i.e. compute loss-function several-times), but which did not help me to understand the underlying ML / CNN principles. A little bit more effort could be done there!

Thx for the course!

Very interesting course , not as the same level of the first two courses BUT it is an excellent resource to get in the Deep learning world. I did not like that they don't teach you how to use tensorflow and keras in a more concise way. And sometimes the content doesn't seem as usefull as the first two courses and more in the deep learning world where computer vision techniques become obsolete so fast...

solid 3.75/5

This course was perfect! I finally understood convolutional neural networks and its popular architectures. Implementing CNNs in numpy was a useful exercise as well. Andrew once again proves to be the best teacher of ML. The reason for four and not five stars is the number of technical glitches throughout the course especially in the final programming assignment. I'm sure they can be improved upon for future classes.

The material is meaningful and instructive.

I did well in a capstone project of image recognition in school with the knowledge from the whole specialization (especially skills on Conv-net in this course).

However, it would be better if we could have access to more computational power and really play with large data set and complicated algorithms instead of just doing simple maths and loading models to see the result.

This course is useful for getting acquainted with the terminology and main ideas that are used in building and training convolutional neural networks, as well as learning the names of a few most well-known existing networks. However, I didn't find the practical exercises to be very useful or relevant -- at least, in comparison with exercises in first two DeepLearning.ai's courses in the same specialization.

A very complete and diverse course with several applications of deep networks. Only problem was the assignments which had a lot of problems in the grader, and created a lot of frustration and lost time figuring out where the problems were. Otherwise it would have undoubtedly been a 5 star course :-)

Keep up the amazing work you're doing, especially for Course 5 where I have great expectations!

Cheers

António