To be completely honest, I loved Dr. Andrew's method of teaching. But the assignments just flew over my head because I didn't have enough hours of practice of Keras under my belt. I know Keras is there to make things easy but it's very difficult to just trying to pass the grader. To goal of assignments was fantastic, I mean, generating music, etc. sounds really amazing but I feel that if there was some more time given to make us better in Keras and other technicalities then I would've loved this course much more!

This course presents a number of ideas that are used in mainstream sequential models. Unfortunately, the lectures are not as detailed and precise as one may wish (and are not on par with the previous courses in the specialization). Also, quizzes and programming exercises are not really useful. In particular, the programming exercises are quite superficial, more of "guess what the test expects" type.

The bottom line: the videos are worth watching (although not perfect), the exercises are useless for learning.

Good but i expected more. The main thing i like about first 3 courses, they were really deep. In the last two courses we have skipped the backpropogation. Now this is something which you can keep optional. I like the way Andrew Ng teaches, going to the basics, and that is why I came here and paid 40 euros per month. Also, there are few stuff missing like Generative models, Adversarial networks, GAN and etc. It would be good if Andrew can have more courses related to this and deep (as it is deep learning :))

The course introduces a lot of new concepts about sequence models. It seems introductory as it only scratches the surface. There are lab assignments to reinforce the learning. I found the lab assignment a bit overwhelming as I encountered new details about the implementation and new reading materials and references while I was working on the clock to finish them within the time limit. Moving those materials out of the lab assignments would give students a chance to review them beforehand.

By far the most difficult of the 5 courses but giving it a lower review since the programming assignments are rushed through to finish 2-3 in 1 week which gets hectic & understanding of key concepts is lost. Also, it would help if more time is spent in the videos to explain the concept/model/algorithm used in the assignments since I close to understood nothing from the assignments in spite of completing them. Finally, the instructions too were not clear in the assignments.

There is too much in this single course. I struggled a lot with understanding the concepts and getting through programming assignments because I didn't feel like the videos set me up well enough to apply what I learned. For me it would've been more beneficial to break up RNNs and LSTMs into one course, and then Attention models and Transformers in another course. I just felt a big shift in the organization and the quality of the content compared to the previous courses.

Interesting and full of excellent lectures as always for Andrew Ng. The programming assignments quality was not as good as the other courses in the Deep Learning specialisation though. They drop straight into Keras with no information/introduction, use several complex model architectures without explanation, in week 3 4 out of the 5 'your code' exercises were about audio sampling, not very relevant. Again, excellent lectures, just not great programming examples.

Sorry to say they dropped the ball on this one. The last course of this specialisation has the most advanced topics thrown at you in just three weeks, and it's even more cookbook-like than in the previous courses. The material of this part of the specialisation would require a whole course in itself, perhaps for +10 weeks. Here, I found it is at best a guide for self-study, _if_ you have the time for that. Also, support in the forums was very minimal.

There are some challenges with the videos eg. repetition, blank audio, variability in speaker's volume (difficult to hear). In particular perhaps 'Bleu score' needs to be redone. I did not enjoy the labs mostly because I don't have much interest in NLP BUT the 'emoji' and 'trigger word' labs were excellent! Especially the 'trigger word' lab should be the standard for all labs, it was very well written: clear, good flow, no mistakes.

The scope of this course was highly relevant to me, but unfortunately many of the class materials were broken or otherwise incorrect, making some ungraded portions of the assignments difficult or impossible to achieve. Activity on the discussion boards indicates many people have tripped over this for at least the better part of a year, but no corrections have been made. This was quite frustrating and wasted a good amount of my time.

I took this course after a long pause after I finished the first 3 courses. I would NOT recommend doing it that way. As a result, I felt rusty on some of the coding practices.

I think the course gives great introductory information on RNNs and LSTMs. The first two weeks of the course are spot on. However, I think the third week is lacking. I had hard time making a connection between the lecture material and the assignments.

This course was at a really high-level and barely scratches the surface of Sequence Models. Didn't really go into much detail behind any of the theory, and the programming assignments were mostly done for us, so you don't really end up learning much. You certainly won't be ready to have a job solving NLP problems after taking this course. If you want that, you're better off going through actual college courses online.

First of all, the programming assignments are really copy-pastes. There is nothing really to storm your brain for. Second, many of the ideas presented in the video lectures are very brief and short, skipping the explanation parts. After taking this course, I now know the names of lots of algorithms and models, but that's all I know, only the names. To get broader knowledge on them, I have to look somewhere else now.

The course content and Andrew Ng are great. The submission process of the assignments is absolutely dreadful. You might get 0 points for correct answers or not, depeding on...well, I have no idea on what. Maybe it's Jupyter Notebook, maybe it's Keras or maybe it's something else. But you must have good search engine skills, since you will most likely spend a lot of time in searching the discussion forum for answers.

May be this is my observation but this is the 1st course where I am unable to understand most of the explanation by Andew Ng, and the course exercises are more like the python coding like slicing, dicing, filtering, and how come this course is same for last 3-4 years, not even objective questions, There has been a tremendous breakthrough in the field in last 3 years and the course content is still the same.

too much information for such a short course. We only get a very superficial understanding of concepts with very little practice to solidify our understanding. The assignments involve implementing very small parts of much bigger systems. I guess the course is ok to get a general idea of the concepts but for deeper understanding of the topics a longer course or multiple courses would be needed.

This is a very interesting topic. However, I believe the course itself can be improved. I believe there can be more information about NLP and sequence models in lectures. It would be nice to add lectures with practical suggestions about training and tuning sequence models. There were lots of typos and mistakes in notebooks that were found by other fellow students and not addressed by mentors.

The course overall isn't as good as the previous 4 ones especially for the PA part, I can pass the grader even with wrong output. The PA improvement sometimes just create more discrepancy. The PA is just a walk through of how to building those basis models, but those little bugs will drain extra hours to figure out. I think this course is kind of a prototype one especially on PA part.

Compared to the previous courses, this was a disappointment. There is not as much content as I expected and the homework exercises are not well prepared. If one spends more time with debugging than with "learning concepts" in a basic course like this, then something seems wrong.

Moreover, in a situation where so many people pay so much money (because of Andrew Ng's credit)...

That was tough, how the weights are stored and their dimensions inside the 'time steps' can be explained by adding one more video, btw the course is awesome if you want to learn the basics of sequence models, you should have completed the previous 4 courses before diving into this. I will always remain thankful to Andrew Ng for providing this type of platform.

Positives : Excellent lecture material. Assignments broadly are well structured. HIgh bar set by Andrew Ng. Negatives: Assignments have too many errors and mistakes as of Jan 2019 (especially but not only in the optional / ungraded sections) for me to be confortable 100% recommending the course. Instructions for assignments are also not fully fleshed out.

This course is good for introduction to sequence networks, but I felt this is not at par with the previous course 4 (CNN). This feels a bit hurriedly done, with many important things only just touched upon. This should have been a 4 week course like the previous module. Then due attention could have been given to the field of speech, audio, sequence domain.

The course is generally good. However there are some issues with lecture videos editing (some sentences are said multiple times), and with activities (e.g. default settings hardcoded in one of notebooks, didn't let have output shown as reference, also in some cases automated grader has some assumptions, which need to be found using trial and error method).

Great course, great activities and really good programming excercises.

I give it 3 stars because instructors let political views tainted week 3 videos and assignments of this course by introducing some techniques for 'debiasing' and making your neural networks more bias to gender equality political views. That has nothing to do with science.

I've got mixed feelings about the whole Specialization. Many very interesting topics, but on the other hands I don't feel like there's any takeaway knowledge for me. Until the very end I've been feeling completely lost in the exercices. I'm proud to have been able to hold on until the end but I'm not sure it's been an useful use of my time.