Worse than the first course - 3 weeks of short and high level content, gains not applicable to the outside of the course. Covering StyleGAN as an advanced architecture does not lift up the shallowness of the course in my opinion. If you are considering to enroll, I recommend waiting until the last course is released because if the first two is an indication, you can easily finish all 3 courses within the 1 week trial period and get a free specialization certification.

Big disappointment! This is not a course it is a seminar, it gives you some idea about some topics in GANs, leave you with a tones of papers to explore while giving little and vague explanations about almost every topic. Intuitive is not equal to unclear!

Not enough useful materal. A whole week (!) devoted to talking about bias, "protected classes" and other social issues as the authors see them... Create a separate course for those who are interested. You don't need to force-feed your religious beliefs, this is plain disrespectful.

Not enough content for an Andrew Ng course

There are too many important building blocks that aren't really being fully explained. The programming assignments should have more, small code chunks for you to complete. I don't want to just fill in the last line of code after the first 200 lines have been written for you. I would suggest following Andrew's MOOC lecture format, with many short lectures, rather than a few long lectures. I would to understand each little topic really well, not get hit with everything at once. I would prefer to have the tests at the end of each topic like Andrew's Deep Learning specialization. DeepGAN wasn't really covered at all. I was a little disappointed with the final lab, after seeing what Style-GAN can do the entire course.

Pros:

*A lot of references to important articles.

*A lot of code in the notebooks that might be useful in the future.

Cons:

*The videos lectures are not comprehensive. This is sort of "self learning" course where one should read the articles on its own in order to really understand things. This is not what I am expecting from an on-line course (and it is also not like what I got used to from the DL specialization).

*Where are the pttx? I want to print them and write some comments

*The "labs" are basically a summary of some concept. There is no added value in writing them in notebook format since the code block is just "lets load this and this, and run".

I think the ethics part (one third of this course) is just too long. It should be way shorter or have some optional parts. It could be great if you put it into a separate AI Ethics course, which many people would probably take, but as it is now, it feels a bit forced into the course.

Nice material, but I think that this course is too superficial compared to the first one. I didn't like too much the notebooks. However, a lot of interesting papers are suggested, giving a good starting point for acquire a deeper knowledge of the topic.

Week2 is little diverged, but concise detailed understanding explanation of style GAN is excellent. It is really worth.

I've just completed the specialization and my thoughts are that everyone should take it (that are interested in GANs! I feel Sharon is a great teacher and the entire team did a really good job on putting togethor these courses. After completing it I definitely have a much better view of GANs, their architectures, successes and limitations, and have a solid background to tackle reading papers and implementing them on my own. Thank you for making this specialization!

With all the positives (which is why I rate it 5/5) there are in my opinion things that can be improved. Especially I think there is too much hand holding for the labs, out of 100 rows of codes I code maybe 2-3%. Many of these don't give much value coding but I want to feel like I did it! Unfortunately now I am left guessing if I have truly mastered the material (and I'm quite sure I haven't, so I will need to re-implement these on my own). Also since you state that calculus and linear algebra are prerequisites then stick with it! You are trying to be too inclusive and there are several part of the courses where I thought it was entirely unecessary because everyone taken Calc and Linalg already has this knowledge. I would prefer instead if you spend this time making other videos where you go in more depth, perhaps going through some of the difficult math etc. Hopefully you try to improve this for future courses done by deeplearning.ai

Too much attention to bias and fairness, but 3rd week is super cool!

Greate course content and assignments but I want to give one feedback to the instructor. Please keep some pause while speaking. She speaks way too fast.

Although the lectures and the presentations of the different GAN models are very helpful and elaborate, the marked notebooks are falling far behind, without actually providing much more than implemented simplistic versions with a few lines of code that need to be edited. The notebooks feel a bit disjointed, for StyleGAN for example, where there is not much coherence in the sense that the participant has to reverse engineer the assertions to make sense of how a fully functional model works.

I want to learn GANs not "fairness in ML"

Exceptional on all accounts. Everybody worked really hard to make this happen. Lots of preparation. At times, the assignments were perplexing but forced me to identify deficiencies. The course is imbued with philosophical subtlety and complexity. It's wonderful to see such a grand push towards programming with responsibility. That advanced spirit has and continues to evade other disciplines. Computer Science is setting the highest standard in excellence, progress, and democratization. The course paints an image of hope for all.

After the first course was to my taste way too simple, this one picked up some speed and is at the right level for me. I guess I wish the exercises were less hand-holding, or that there'd be an optional lab to build and train a full StyleGAN, where the curators would help with the more technical questions like linking a clean dataset or setting up a data loader. Looking forward to the course #3

Another beautifully clear course. In particular I enjoyed FID and found A Survey on Bias and Fairness in Machine Learning an interesting read. The assignments, as always, are enjoyable, and through the unit testing give a more practical understanding of what is going on.

Very good course! Helpful to understand evaluation metrics and details of Style GAN. It was also super cool to have the bias section that is not as well known as the others. Loved it!

Great material...but the stylegan code implementation requires more video material. Instead adding one more week for ProGan part before stylegan would be helpful for the learners.

Week 2 could explore more into GANs variants, StyleGANs or VAEs rather than focusing only on Bias theory. Overall very amazing course. Week 2 can be improved, I think.