Assignments are not always consolidating acquired knowlegde, due to the distance between implementation from scratch and using predefined models.

Very nice course.

assignments are very high level though and just help in giving a taste of convolutional neural networks.

lecture notes are great

A bit buggy some of the exercises, some of the videos and some of the course notes, but the material is excellent and the learning is invaluable.

Amazing course that breaks down the complexity of CNN

Some assignment have issues yet the forum displays significant resources to help solve them

Some key points for not giving 5 stars (lack of questions during lecture, assignments too easy). But, as always Andrew is an excellent teacher!

The course is complex, it must teach more examples of coding and practice overall the content of this course requires very high level practice.

I really like the videos, and the content covered. It would be nice if the large blocks of videos were separated by some programming exercises.

The coding assignment is a little hard for people who don't know much about tensorflow and keras. But on the whole, this is a brilliant course.

There is a great course for many people to begin working with the such an enigmatic stuff as cnn! Thank you very much for creating this course!

Wonderful and well-structured course! Following instructions on jupyter notebook is confusing, had to go back and forth to do the assignments.

Grading assignments troubles are annoying especially after such a good teaching it is really frustrating. nevertheless it's an amazing course!

Interesting course, but the assignments were sometimes disappointing. There were bugs/issues, and it wasn't very clear how to deal with these.

The coverage of the material is excellent. The only downside is that there is little support for the exercises and can be quite frustrating.

I think perhaps you can add more detailed explaination on Fast R-CNN and so on. Also, I think the homework can be updated to Tensorflow 2.0.

It's very difficult. I believe it can be more simple. But on the other hand maybe when I go through it 2 or 3 times it'll became more clear.

the labs need to be little bit better with more code to write and better explaination in the later courses because i am still struggling :(

Andrew could have explained a few concepts in depth overall a great course though giving an overview of CNN's & deep learning architectures

Pretty good course, holds a lot of relevant information.

Though the going into every little detail of doing a convolution is a bit overkill.

I think Keras should have been used more in assignments. Overall the concepts are illustrated in the videos perfectly as well as precisely.

Instructions were clear and videos were well thought out. 4 stars because I had to reference the Documentation for Tensorflow consistently

this course had all the practice exercises and todo assignments with proper lectures and examples to understand even such complex problems

course is good, homework has bug in week4's homework, meeting expected output is graded wrong, changing to different output got correct.

Errors in grading procedure. In many lecture videos you hear that they are consist of many recordings that do not compose one whole thing.

Would have given it 5 stars but the notebooks were problematic, especially week 4 Happy House. Otherwise, it was a good conv net course.

Course content is very good and I found useful but I think Yolo exercise does required some changes. Over all very good, Thanks Andrew :)