Excellent Course on NLP. Need more focus on RNN/CNN/Sequence modelling

The course is very nice. I wish there were some more examples in slides to understand the working of algorithms. Maybe its advance that's why I felt it.

If notes of every week were available then it would have been very beneficial.

The course was great. Perhaps it'd be nice to combine the lower level with some higher level Deep Learning framework at the end. Thanks!

Overall great intro to NLP. Basic techniques like word embeddings and attention are explained quite well. However, some topics are not really easy to remember from this course, such as Topic modeling using LDA (which I understood much deeper during Bayesian Methods for Machine Learning course from the same specialization).

Course is well organized. Just some problems are related to assignments, There no exact guided steps for assignments and quiz. Lectures are more concerend about theory and less about pragmatic problems. Please make discussion forums active and more connected. Course does not have option to send message to anyone personally take help and advice which is something unrealistic.

This course offers a really good intro to all of the state of the art techniques used in NLP. Its course is structured by heavy impact papers from the literature and the instructors do a really good job in explaining.

The quizzes are good and help you understand the material.

If there is one thing I didn't like in this course, it was the programming assignments. Their structure was really big and aimed really long. Thus, I often felt that I didn't know what I was doing and for what reason. The assignments from Andrew Ng's Deep Learning's Specialization "Sequence Models" course were far better and helped me gain much intuition on how to code real tasks.

It was a good course. All in one breath. I would like to have more practical tasks in which you need to write more code yourself, and not just fill in the missing cells. Of course, I am very grateful to Anna and Andrey for their work. Separately, we would like to note the penultimate task-the final project. It is complex and interesting. I believe that it would be correct to use the computing power of Azure instead of AWS.

The quality of this course is good, but I still feel that the algorithm introduced is basic. It is better to combine the leading edge algorithms. Actually, I finished it very early, but I stuck in the fifth week of the dialogue robot job for a long time. I feel that this part of the time is not very worthwhile, because time is spent on the more cumbersome model deployment instead of focusing on Algorithm optimization, of course, model deployment is also very important, this is also a test for me, so I give 4 stars.

Good information, but I am struggling understanding the presenter. The course feels a bit forced.

This was a good class, and I want to give it more stars, but the default path for completing the final project requires far more time than the course specifies, making it difficult to properly manage one's time.

Good and useful content but weak quality in providing directions in assignments and zero help

The course material overall is good. However, the assignments are not very beneficial from my point of view.

The peer review is totally a mess -no one really review an assignment-

This is a very good course, brought with clarity and humour, and excellent quizzes and assignments to assess your understanding of the material. The big problem it has is the final project, which requires you to run a piece of software on an AWS free tier server. An impossibility, because the software requires more memory than Amazon offers in its free tier. Moreover, as it crashes every few hours, peer reviewers can find it not working and have to contact you to restart it. Constant monitoring for several days is required.

Shallow theoretical explanations (just mentions and references). Disconnected assignments from theory.