Great course! Classes of Andrew Ng are, as usually, crystal clear about necessary theory and full of precious hints for efficient implementation of CNN. I recommend it to everyone seriously interested in Computer Vision advanced tasks.

Hardest of the 4 so far. There's more autonomy required in programming and shape calculations require really understanding how ConvNets work. But the more difficult it is, the more worthwhile and non-trivial the achievement becomes. :)

Excellent overview of CNNs including practical exercises with appropriate level of details. Gained good understanding what one can accomplish with CNNs and where to start. Also gives good idea of practical implementation costs of CNNs.

It was such a great course. I got a chance to study and apply fundamentals of ConvNet to many practical problems such as Face Recognition, Object Detection, Semantic Segmentation, Neural Style Transfer which I have longed for knowing.

The course is really good with in-depth explanations of the concepts in a clean, clear and precise manner that is both easy to understand and implement. The programming assignments are fun to complete and test out. Highly Recommended!

I found this to be an excellent introduction to convolutional neural networks. I was already very familiar to convolution but I could see that if I wasn't it would have been clear. All concepts were explained well and I learned a lot.

I think this course is excellent and I'm already applying the skills I've learnt from it to my current research. I would have preferred a little bit more focus on the theorical part of ConvNets, especially backprop. 100% recommended.

Very clear lectures and hands-on experience to gain lots of experience with CV problems and cutting-edge models. I'm an NLP engineer and this course gives a great intro to DL for CV. IMHO it's one of the greatest course in the series.

Thank you very much Andrew Ng. The course helps me understand very well about Convolution Neural Network. Lectures go from easy and then combine together, even assignment, I feel very excited and happy to learn in the teacher's way.

This course is very good if you want to enhance your knowledge about CNNs, the course contains different CNN architectures and use cases, the way Andrew Ng Sir teaches every concept in detail along with visualization is appreciable.

This is an immensely helpful class. I have been wanting to learn imaging processing and machine learning, and this class helps me get started. Using what I learnt from this class, I was able to implement CNN to help my own research.

The most interesting course in the whole deep learning specialization, a lot of practical cases and much closer to the deep learning state of the art. Kudos for face recognition and neural style transfer (yolo is super cool as well)

Fantastic overview of CNN's from a historical perspective but also from a practical and theoretical perspective. Basically all facets of CNN's have been very clearly discussed. Andrew Ng is just a pleasure as a teaching instructor.

This is a great course which make me know how to do computing vision and neural style transfer (which is something I thought amazing before). Although the course is a bit difficult, it is interesting and useful. Hope you enjoy too.

I found this more challenging than the previous courses, but learned a lot. The concrete examples of CNNs were really helpful to picture how they are built in practice and the exercises helped me understand tensorflow much better.

Week 3 was most relevant to my work on Indoor Positioning System. I felt like week 4 could have been made optional. Other than that, excellent material, easy to understand and digestible for us students from engineering discipline

Wonderfully explained! Andrew and team have been kind enough to provide all the important papers and documentation required too. Very well laid out course. Can't wait to finish the 5th and final course! Thanks team deeplearning.ai

An optional course on Tensorflow/Keras and their comparison with other prevalent frameworks would have given a nice touch. I realized that lot of handholding is needed in assignments just because the basics if TFlow are not clear.

Excellent course! And the programming assignment is necessary if you want to know deep learning deeply, the video is shallow and it is more like an introduction to deep learning . Anyway, Andrew Ng is absolutely a great teacher!

This course offers quite plentiful materials. I learned lots of models whose performances are state-of-the-art. Brief and intuitive description of these models helps me a lot when reading the corresponding research papers. Thanks!

I love how the content of this course is structured. Also love the fact that all the weeks contents can be found within the exercises themselves. Thank you to all the people that worked so hard to deliver this exceptional content!

A very well presented course that covers a good breath on different types of CNN model to date. The exercises are good, given the computational limitations, it is understandable why most of the exercises used pre-trained weights.

A more consistent Keras or Tensorflow workflow would be good. I passed all the assignments but still feel unclear abotu Keras and TF workflow. But that will come with practice. Great teaching in terms of conceptual understanding.

The course is really helpful to beginner. However, I really suggest to add more introduction of modern CNN networks to this course, such as R-CNN for semantic segmentation. Really appreciate Prof. Ng and Coursera for this course.

Step-by-step convolutional networks are presented. It is excellent learning to construct the convolutional networks in the lab. I read too much about these type of neural networks, but no one has me shown before how to build them