I gave 5 stars to other courses in this series but this one doesn't deserve 5 stars. There were many typos and bugs in the assignments compared to the other courses of the specialization.

Video Lectures were excellent.

Assignments, however, were buggy and spoonfed you too much. I completed them, but on more than one occasion, I had no idea what the code was actually doing.

The lectures are wonderful as usual but the practical assignments are so extraordinarily perplexing that it does not seem to improve the learner's comprehension by a significant extent.

Boring course:

Short videos instead of lections.

Copy-paste task instead of exercises.

Loading pretrained models instead of training.

Good full keras tutorial will be ten times more useful.

The lectures were not as good as the previous andrew ng. courses, and the exercises were quite bad in all honesty.

I do appreciate what I have learned, as the lectures WERE clear enough.

Just one observation, the general view I have from this course is very positive but compared to the other courses the program exercises in this course are more difficult to understand.

A good introduction in Sequence models. Unluckily the programming assignments are more of a guess work then real knowledge as there are too few possibilities to check the own code.

Despite being a great introduction to many NLP concepts, the programming exercises are a bit too much "fill-in-the-blanks" and not challenging enough. It is still a great course !

Course contents were very good, but the assignments could have been improved a lot. Couldn't get much value out of the assignments rather than blindly following the instructions.

The class covered too less on the Sequence models such as LSTM and GRU. It has too much domain knowledge related videos. Also, it does not have any video related to Time Series.

The videos are good, however the assignments are too long with sooooo many text inside. You have to make it more efficient somehow.

But thank you anyway for the great content!

Assignments in this course are relatively less organized. You have to read a lot of description first and then code and have to go back/forth between code and description.

I believe that the course needs more time allocated to the incremental teaching of this rather large subject area with varied applications. Just needs to be a better way.

The course was especially interesting in week 2 and 3, but the assignments for week 1 were confusing and sometimes it was hard to follow where the logic was coming from.

This was the most difficult to understand course in the whole specialization. Would have enjoyed more if the course material was a little more spaced and elaborated on.

Practical Assignments with Keras wasn't motivating. I spend more time to deep into the Keras rather than into the course topic. I prefer them to be using TF or Python.

The supports in keras programming was so poor, that I could not quite understand each step. And the server was horrible, always got 405 response when saving my codes.

This course should be broken into 4 weeks and spend more time building skills with Keras. The number of hours listed next to each assignment is unrealistically low.

The lecture was great but the errors in the programming assignment (especially in formal-typed formulas) really wasted a lot of time and make me confusing at first.

The ideas presented here were clear, however I found the programming assignments non-intuitive and not practical. I spent on them way more time than I wish i had.

I enjoyed all the courses, from my personal point of view this course was not that fun as the other courses. Except for the trigger assignment it was awesome!

Too much was given in external python code for the first week's assignment (that should be learnt by us): not a good thing for us to gain a good understanding

Worst course of the specialization, content is interesting and Andrew keeps explaining really well but programming assignments are clearly of a lower quality

The course is great but the tasks in Keras are too complex without background knowledge. Therefore, a reasonable introduction in Keras would be desirable.

Far less detailed than the other ones. The programming assignements are less interesting too, as a great part of the work consist of reading documentation