I think the transformer programming exercise of this fifth course is not as good as the others. The methods we must implement are not clearly explained and the research on these really took me a huge amount of time.

The exercises are too short and too basic for this course specifically. The task is a great learning experience but there is not much one would struggle with in terms of difficulty as there is too much spoon feeding.

Great video lectures, but practical assignments are a pain due to awful auto-grading system and programming expirience in Jupyter in general. Most of the time you'll be searching for an error that isn't really there.

Excellent course, excellent lecturer. Unfortunately some of the test data (week3/lab/trigger word detection/XY_dev/* CANNOT BE DOWNLOADED ... The programming lab sections are nice - sometime a bit too helped ... ;)

It is better to talk more about bi-directional RNN and give assightment about it before week three and week four. Moreover, it is also better to give two more weeks to talk more about the material in week 3 and 4.

Programming exercises did show you quite a bit, but got complex enough that most of my time was spent reading and understanding the preamble than doing any programming. That being said it delivered on the promise.

Course content is really good! However, I found it quite difficult to truly understand deep learning algorithms. However, it provides good glimpse of of sequence models and intuitions behind various useful models.

Great course and material, Andrew NG really know who to explain difficult subjects in an intuitive way. However, the course seems that it still needs some work (there are some bugs in the lectures and assignments)

The lectures are suitable. Andrew, as always, is very clear and goes into necessary detail on each topic.

The programming assignments have adequate assistance and hints, except for the final week on Transformers.

The NLP Section of this course is quite difficult to understand(The Notations are quite confusing as well as prior knowledge is required to understand) but other than that RNN, GRU, LTSM are explained clearly.

Week1 lessons are a little complex than the previous classes and there are gaps between ground-up python version and keras version of LSTM model. Keras will need to be taught a bit more in detail to follow up.

This is by far the hardest course in the specialization. But it was explained well. My only complain is there were errors in the first programming exercise. All in all, I learned a lot in this specialization.

Great content! I really love Andrew's teaching style. (1 star deduction for some programming assignments where I spent time debugging but it turned out that the point deduction was due to the grading system.)

Great course, but the transformers section was much, much more difficult than the rest of the material. More time, lectures and exercises should be given to give us intuition about these very alien concepts.

Overall course content is great but I feel like improvements can be made on the lab exercises to give students concrete understanding along with hands-on experience. A great job by the deeplearning.AI team.

The programming exercises are a little heavy in this course where we need to load and re-load for them to give correct output even if the code had been correct all throughout. Otherwise, the course is great.

4 stars and not 5 stars because the course is shorter than the others and it feels like an exemple in classical forecasting is lacking (sales, time series...).

Really interesting but may be too focus on NLP.

It's an excellent course! I will give it 5 stars if it could offer more interesting and meaningful assignments(Not offend, but it a little too easy and the assignments are not very related to the real work).

The course should contain more explanation about natural language processing like tf-idf,lemmatization,stemming,dialog flow. Although i got a good explanation of working of RNNs,LSTM and machine translation

Great course to start off with sequence model. The programming exercises were in depth and deliver a great learning experience. Would love to see more of sequence literature in the course's future versions.

Again a brilliant course from Andrew NG, but though and dense this time. In order to grasp the meanings videos and lectures need to be revised a few times. Also, get some extra info elsewhere does not hurt.

Excellent presentation, and interesting assignments. One star dropped because a couple of technical issues with the assignment material (typos in the mathematical formulas / expected results here and there)

The contents are great as always. However, the server is not reliable. Once, the grader is down and you can't submit homework. For another time, the connection is lost and all the changes made are lost.

Great course, but I felt the gap between the very high level lectures and very low level labs was a bit too wide. I had some issues with the autograder and losing progress in the notebook between sessions.

Great course. The only frustrating part are the programming assignments as it is very cumbersome to have to go thru the discussion forums to resolve issues. Tough to find helpful insights thru the forums.