Exceptional Course. Learnt a lot from it. Takes a different approach to teaching than other courses in the sense that more focus is on applications rather than training of models for which a GPU cluster is a must. Thanks Andrew Ng and his team for the wonderful course. Looking forward to sequence models :)

An amazing course to get an advance knowlege and practise "Convolutional Neural Networks". Andrew Sir makes the math and concepts behind the scenes very easy to understand. The course is easy to follow as it gradually moves from the basics to more advanced topics, building gradually.

Highly recommended.

Professor Andrew is really knowledgeable. The lecture videos he makes are really helpful for me. I really learn a lot from them. Also, the recommended learning materials such as academic paper he recommend are really useful for me if I want to further my learning on the residual network or YOLO algorithm.

Great course by Prof Ng. I had taken his Machine Learning course a few years ago, so expected high standards of content and assignment preparation - I was not disappointed. Staff is very responsive and helpful in forums as well. I highly recommend it. Taken as part of the DeepLearning specialization.

Exceptional course taking you into the real world of deep learning by exploring new concepts and classical architectures like LeNet-5, AlexNet, VGG-16, ResNet, R-CNN, YOLO, FaceNet and Style Transfer that propelled deep learning in new heights. Loved every part of it (minus some hiccups with the grader).

The course is just awesome both in terms of content that is being taught in the lectures and the assignments. Though, I think the last week is not that much important for the industry purpose but definitely it is good for those who are interested in non-industrial use of tensor flow and neural networks.

I finally understand YOLO! This course has the best material available on CNNs. Even though I come from a MSc in Computer Vision and Machine Learning, we didn't have enough time to fully cover 'complex' architectures such as YOLO. Thanks to this course I feel more up to date in the Deep Learning field.

Once more Andrew steps up as a brilliant teacher. I'm a biologist looking to improve my data science skills to better tackle medical imaging problems. I'm confident to say Andrew is the reason I'm going to make a difference in low resource communities in the future. Thank you, Andrew, you are awesome.

I really enjoyed this course. I found it pretty approachable. FWIW, I'd taken Andrew's original ML class, but then skipped 1,2, and 3 of the new one (and jumped into 4) The course really holds your hand, so be prepared to force yourself to try some of this on your own to be sure you've understood it.

This course is intended for ML learners who have background knowledge of NNs and want to enhance their scope of knowledge in CNNs. Prof. Andrew has been an amazing instructor. The material used in this course is mostly based on Tensorflow, so make sure to have a bit of prior knowledge in Tensorflow.

I have learnt a lot new things in this course, constructing exciting image/object detection projects with Tensorflow, Keras and even plain Numpy. Also, Andrew well explains many complex network architectures which illustrate various perspectives of the applications of convolutional neural networks.

This course really gets you started working with CNN. The only downside are the "bugs" in the assignments. My advise is to read the discussion forums before you do the assignment to know if there is a bug that you should know of before submitting.

Beside this minor bug, the course content is 5 star.

Really enjoyed learning more about the current state of the art of image recognition models. Although the structure needed can be at times overwhelming, the concepts are clear and implementation via open source packages make it feasible. Many thanks for making this available, keep the good work!

Somehow, a bit harder than rest of the courses for me. I had problems with tracking dimensionality and tensorflow notebooks were hard and difficult to debug. I think it would be nice if tensorflow has its own as a course or 2 weeks maybe. But anyway the concepts explanations is great as always!

Excelent course. ConvNets are an eye-opening subject and the course explains the main concepts and applications in a simple way, indicating the source papers to understand better. I'd only ask for a couple of videos explaining in more detail backpropagation and the upload of the missing slides.

The content, examples, assignments, and quizzes are thoroughly developed. All the courses of the specialization share the same notation and lead a student from basic concepts to complex ones helping to develop an intuition on each step. The best course on topic of Deep Learning one could find.

great ! It is complex though, don't get fooled by the doable exercises - to really understand you can take several loops.

Imagine someone breaks up recent complex research paper into python notebooks for you and you get this delivered like a delicious food - this is how I feel about this class.

This is an excellent course. Although I've got 100%, there are still some details and intuitions need to be figured out. Maybe I will go over it again. And of cause, I'm looking forward for the fifth course. I wish the fifth is not the last course. We still need to know reinforcement learning.

Very informative lectures with simple explanations of what the algorithms are doing. The programming assignments are extremely detailed and well explained. This makes it very efficient and fun to learn the concepts of Conv Nets, Res Nets, the YOLO algorithm and so on in a short period of time.

A great course overall. Ties together the concepts presented in the first 3 courses and does a great job of showing some very practical real life applications - the programming assignments show a wide range of practical applications of deep learning like face recognition, art generation, etc.

It was a great course. You end up with a pretty good understanding of convnets and their different applications and algorithms. For sure this course set up the basis for image processing work and research, although it is necessary to refresh concepts and go over the notebooks to fix concepts.

The course explains topics I used to consider as "complicated" in a very clear and simple way. The videos and quizzes about theoretical concepts accompanied with programming assignments and extra reading material give solid understanding of the topic, its current trends, and future direction.

I think that should have an optional video with the mathematics behind the convolution/cross relation, showing element-wise operations on a small volume with more than one channel. I know most people will find it boring, but i think it will make easier to fully comprehend the 1x1 convolution.

Really enjoyed learning this course. I'm a PhD Student in CS but neither in computer vision or NLP. I feel like these courses are sort of jump-starter, if you would like to learn more about DL and to be expert, there's a long way to go. However, this is really a good starter!! Thanks Andrew!

The best place to start Computer Vision! You'll get to implement state of the art Techniques in CV, most with practical Application. The quizzes are very well designed and test your concepts. You'll learn to use open source implementations and build on top of that as well. Wonderful Course!