Too much focus on YOLO and other very computer-vision specific applications. The general introduction on ConvNets is good, but there are other applications than stuff for self-driving cars. I wish the examples were more diverse. In addition, the Jupyter notebooks used in this course are extremely unstable. You're unable to save your progress, and there will be problems submitting your coursework.

While I enjoyed Andrew's course on NN, I am a bit disappointed with his CNN section for one major reason: he did not explain the philosophy behind filters, etc. Instead, he tried to cover too many things based on the latest developments in the field of CNNs. Take this course if you don't mind being exposed to the subject without understanding deeply (no pun intended!).

The course itself is great, but the grader is seriously broken and the staff has not been willing to fix it for more than a month. So basically when you are finished with correct implementation, you spent lot of time in frustration trying to get through the grader, then you go to the forum and find out what need to be changed in your solution to pass the grader...

Course material is informative but when asking for help grading they just point you back to the discussion forums (i.e., no help). Having the output of the grader actually show what your output is would be helpful rather than just saying, "wrong, try again." This is especially true if your output in the assignment matches the practice example. Very frustrating.

The first two sessions are very well explained, with clear and precise examples. However the last two sessions, are explained in a very superficial way, without a good example, the explanation of these sessions are not deepened, the practical exercises don't teach how the problem is really solved. To truly learn, it is necessary to go out searching the internet.

The content is good, but the videos have not been edited properly. There are spoken notes to the editors left in the video and you can hear Andrew rephrase sections. To me that is really distracting, as I now start thinking about how Andrew phrases things, instead of about machine learning.

So a good beta. but certainly not a finished product.

The subject of the lectures are good. However, Andrew's voice is still relatively small with other noises.

Also, there are quite a lot of times when same words are repeated in the audio.

Finally , during the lecture a felt a little bit of sinocentrism from Andrew.

Multiple of the videos have editing issues and repeat clips. Programming excercises were good, but final programming exercise was a pain to finish, not cause of difficulty but cause of having to debug code without any proper feedback on whats wrong.

The lectures were taught far above the heads of my colleagues and I and the practical exercises were far too simple. I really wish the instructor took a course on pedagogy or went through Software Carpentry instructor training.

Week 4 videos were not edited at all. Week 4 lecture slides were not available for download. Week 4 programming exercise grader had significant errors such that the incorrect solution needed to be coded in order to pass.

Some of the topics were not explained in enough detail and felt like being quickly skipped. There were some problems with the grader system in one of the assignments which wasted a lot of time and caused frustration.

Content was great but a very buggy grader in week 4 made for a stressful experience that upset many students. Grader bugs caused me to repeat the course in another session and those bugs remained in the next session.

Programming assignments did not deal directly with the CNN models, but with auxiliary functions. Hacking the grader was more important than getting the right answer. Videos should have been better edited too.

Lectures are awesome, really inspiring and intuitive.Trouble with submitting assignments. I've solved the given question and resubmitted for almost six times, but the kernels showing up errors.

Horrible user experience with the "Jupyter Hub" constant issues that makes trying to do the exams an absolute nightmare and a perfect anxiety booster!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

In some tasks the delivery format is arbitrary and does not specify well what is wanted, that is, so the numerical value requested is good, the output is incorrect. due to format issues.

Programming part was not explained well. I guess programming syntax and flow of code should be explained too instead of just telling theory or focusing mainly on theory.

TOO MANY BUGS IN THE EXERCISES.It was a dis-incentive. Really.And I love me some deeplearning.ai! It has been great. The videos and content structure are fantastic.

Though the videos were very good but the assignments require too much from us and also there are few mistakes in week 3 and 4 notebooks which take a lot of time.

https://www.coursera.org/learn/convolutional-neural-networks/programming/IaknP/face-recognition-for-the-happy-house/discussions/threads/NcpP7i95EemJswr-eOHMNg

The content is excellent, but there were technical problems with the final homework assignment that were not addressed by staff in a timely manner.

Content was interestind and provided good theoretical overview. Exercices where you just have to fill in some line of codes are not usefull.

Some of the code was incorrect and the guidance was often confusing. Visibly worse than the other courses in the specialization,

Don't feel like I understand these models much better than before. Still don't see the logic of the identity layers

Lots of problem with the grader. Wasted a lot of time grappling with grader issues. Very disappointed.