Basically a shallow introduction to programming simple CNNs with Keras. A lot is reused from the first course in the specialization. Reading one of the Tensorflow Tutorials/API documents on CNNs, Dropout, and TransferLearning will be time better spend, than doing this course.

The course is Ok, it gives you some insight on CNN and some useful tools in the Keras API. However it is quite simple and it doesn't explain the fundamentals behind it. The final tests are very simple, but can get quite complicated if you don't attached yourself to the tips.

The course itself is really good; however, homework problems at the end of the chapters are very unorganized. There is almost no guide at all. You may end up spending hours while trying to figure out why grader is having problems or your model's accuracy is very low.

Giving it 3 stars because of the last week's assignment. There is little to no information about the dataset and the learner is just expected to know how to deal with the data. No information on how many classes to expect as output and other necessary information.

What could improve it: Not enough depth in the practicals if you have already done Andrew Ng's course on Conv nets. No graded practical exercise.

What was good: Clear examples, Good setup to experiment with the algorithms & Speak explains concepts very clearly,

Good course, but the notebooks need a deep review to fix the problems related to balance between the requirements of the exercise and the resources available also a better explanation of the exercise aims would be a nice to have to avoid misunderstandings

Actually a great course. Only not getting more stars due to the issue encountered with the last exercise where there is an issue in loading the data files. The workbook keeps on crashing and there is no solution provided to resolve that.

The material is good, but there is not much thereof.

The duration of the assignmentsis greatly exaggerated, since most of the lengths for the readings and exercises are wrong.

The course can easily be done in 25% of the official time.

If you have taken the first course of the specialization this class was repetitive at some points. I also did not like that there have not been graded coding problems. I still got some practice and learned some new techniques.

The materials about implmentation of transfer learning is helpfu, but again, I think the whole content of the first two courses could be compressed into one week. There're really not too much new things.

some clear instructions should be given for students. In exercise of week 4, there was no specific instruction about using .flow instead of .flow_from_directory, for labels we had to use to_catagorical.

Course assignments need a good overhaul. The two environments to practice the assignments (Jupyter workbooks and Google Colab) are not consistent, one throws an error while the other one works fine.

Very good course but please consider reordering the videos and reading especially in week 3. It is better to discuss the code in the video before moving to the notebook not the opposite.

Thank you

Course content was super nice.

But exercise organization is very annoying. not at all satisfied with the exercises. sometimes not loading and sometimes is really annoying . very disappointed

Good content for classification tasks. But didn't cover anything related to object recognition, localisation and semantic segmentation which are the challenging computer vision tasks.

The essential of convolutional neural networks is covered by this course although there ais unnecessary code in the examples and a lack of explanations especially in the assignments.

Too basic course. If its a practical course, then there should be exposure to more functionality of keras and not just the basic one which can be done from a blog/documentation.

Lawrence's lectures were good, but exercises were disconnected from course material. Having to do exercises in Google Colab and then redo in Jupyter notebook was inefficient.

Too easy. One can finish all exercises without learning much. The quality of explanations is poor. The whole course is but a short walk through Laurence's Jupiter notebooks.

the topic is interesting, and the course is quite hands-on, but the treatment of the subject is extremely basic. Videos are too short and somehow superficial and incomplete

cnn implementation theory should be covered better, giving more reason why the code is written this way, furthermore the last homework of the course was bad designed

very good course, but think it needs to go deeper in the functions and tools in tensorflow for conv netwroks, i have the feeling that the course is somehow shallow.

It was a great course, but in my opinion, it could have been even better if it involved more concepts & APIs to explore apart from the most in-use TensorFlow APIs.

Solid content, but it feels like it's not *very* much on top of the first course in this specialization. I think these two courses could be combined into one.

The course is good for beginners as it is very basic. It needs more advance topics like Detection using TensorFlow. Have a lot of scope for improvement.