I am disappointed by the depth of material that is presented in this course. I could easily find the same level of content by a simple web search on "intro to gans" and gained the same knowledge by reading the 2 medium articles. Lectures and exercises felt shallow and no different than simple kaggle tutorials. I feel like deeplearning.ai course quality is getting worse and worse with each specialization

I usually do not write reviews but I think I can save other's time and money by doing so. In my opinion this is one of the worst courses on deep learning.ai. The irony is that it's still the best among the the three courses of this so called "GANs specialization". It's more like a one day seminar in a big university, it gives you a vague idea about the GANs, zero or practically non existent programming experience on GANs (since you basically fill some lines in a prewritten codes). Also, the instructor really needs to improve her teaching skills and to speak slower. This was not worth my time and money, I am sorry.

Instruction is awful. Bulk of time in videos is spent with high level, executive style summaries. Any time spent with actual details pertinent to the lesson is often done on a single slide where one has to bear with the instructor as little lines and symbols are scratched back and forth across the formula or diagram of importance.

The course is great and the assignments are informative. However, I have a couple of suggestions: 1) While Sharon did a great job explaining things in the video, I think speaking in a lower pace will get the idea across easier. 2) Sometimes one might not follow what the lecturer is pointing at in a slide since there is no pen input present. I suggest more pen interaction with the slides. 3) While concepts are present in the videos, more than often the details are left to the explanations included in the assignment. I think it is better to include all related details of a concept in the video as well. Overall - Great Job Team!

After having greatly enjoyed the Deep Neural Network specialization and having learned a lot from it, I had great expectations for this course. I have the impression that I did not manage to learn as much as I could, as most of the topics were covered very briefly and without giving enough details. The notebooks were also quite unsatisfactory. Having no prior experience with PyTorch, I was struggling at points, e.g. because I did not know how certain function behaved or how PyTorch is structured. Most of the exercises were either too simplistic or too cryptic for me. While I fully believe it is necessary for us to go over some stuff on our own, I feel that the type of exercises proposed during the course were mostly a coding challenge (e.g. find how to calculate the norm in PyTorch), rather than an explanation and a hand-on tutorial on how to build a GAN from scratches - 90% of the code was already there.

The lecturer talks too fast and in a very repetitive manner.

I have been wanting to learn about GANs for a while but this is the first course that breaks it down block by block. The course offers insightful theory coupling with practical examples and exercises and close guidance.

Sharon is doing an impressive job providing introductory explanation to the GANs. I am the type of learner who prefers to learn something through systematically organized contents and this course really suits my learning style. It provides basic overview to the current trends and the applications of GANs, as implied by the course name: Build "Basic" GANs.

Well done!

great course, only teaching what's needed, doesn't push you a lot in the coding assignments, as much as it requires you much more work to understand the codes and the science behind it.

This course was not at par with the other courses that I took from DeepLearning.AI. The course content is quite complex, but the way it is unfolded to students doesn't level up with the complexity. The programming assignments are almost done for you and there is not much that one can learn from them. More end of session quizzes and code practice can be added to drill a more thorough understanding of GANs.

A wonderful clear introduction, the basic concepts are presented in a manner which even someone like myself could understand. The exercises are helpful in forcing you to play with the mechanisms involved, are well documented and are not overwhelming.

The course provides good insight into the world of GANs. I really enjoyed Sharon's explanations which were deep and easy to understand. I really recommend this course to anyone interested in AI.

It is good but this is not up to the mark like professor Andrew ng. detailed explanation like him is necessary.

Another thing is mam is talking too fast. she should take a break while explaining the things.

Videos were vey short with very brief explanation.

Excellent course. The videos were a pleasure to watch, the assignments were clear and allowed you to go as shallow or as in depth as you desired, and the mentors were very helpful.

the teacher is speaking too fast...

The idea of the course is great, but the realization is TERRIBLE!!! :( That wierd chinese schoolgirl mumbling 'bout kitties and golden retrievers insted of giving instruction on programming is just annoising!!! The tasks are completely diverse with the lectures and very demotivating. :( I'm really disapointed after this course. :( Conclusions: try this only if You're ready to spend a lot of time googling the answers and cursing.:( WE WANT LAURENCE!!! WE WANT LAURENCE!!! WE WANT LAURENCE!!!

Loved the course. The lectures were exceptionally clear and I learned a lot about the most recent developments in GANs. The homework was the weakest element. Adding a few lines of code to a routine probably isn't the best way to test your understanding of the material.

The course is amazing with an amazing instructor. I really enjoyed the course and thank you so much for making this specialization. A big thanks to deeplearningai team.

This course has been long waited for! It is great addition to the AI community and it presented very clearly. A bit of more theoretical background could be helpful.