The instructors left out too much on the intuition part.

Well paced and easy to follow.

There are some typos here and there (so the course might need some more polish on that end) but, apart from that, it's accessible and puts the focus on understanding the concepts rather on coding contrary to what I have read in another review.

You could follow the course even with no prior experience in Python.

If you take the course, don't skip the ungraded assignements, they are an integral part of it and provide more detailed explanations of what has been taught in the lecture videos.

Thanks to the team and good luck all.

Great Course,

Very few courses where Algorithms like Knn, Logistic Regression, Naives Baye are implemented right from Scratch . and also it gives you thorough understanding of numpy and matplot.lib

A really great course in NLP. They do a really good job balancing beginner and intermediate skill levels. This is a good introduction to NLP and machine learning in general. Really fun course!

Fun, interesting and useful course. A couple of road bumps in the assignments made me waste a lot of time, but the forums and Slack channels were lifesavers in those situations.

I has the right mix of challenge and support. I gained new insights into topics that I thought I already understood well. Great introductory course.

Very helping in understanding the maths behind NLP for classification methods and I can see these things more intuitively from now on

Good job! The course material is easy to follow and the links to related material is appreciated.

Great course, very good materials and explanations! ❤

General Comments on Course 1All of the linguistic and semantic knowledge that we were mining in Course 1 was encoded in the vectors. The coding was just using statistical methods to draw linguistic inferences from these vectors. I find it unfortunate that we didn't have an opportunity to learn how the vectors themselves are made (or did I miss something?), but merely got them out of a can from Google.It took me a while to figure out that the homeworks are graded by an AI, not a TA, and that one can submit the homework assignments numerous times until getting the grade you want. Given the large number of students enrolled, I can understand that hand grading would not be an option. It would probably be helpful to explain this to newcomers like me.I liked the way the assignments are structured. the fill-in-the-blanks approach, followed by some sort of numerical unit test to let us know if our solution was correct is good pedagogy (androgogy?). My only criticism is that sometimes the unit tests are not very sensitive to common coding errors. But now I know that we always have the option of running our assignment through the autograder for more complete feedback.The autograder is often overly prescriptive. For example, "Function 'np.sign or np.heaviside' not found in Code Cell UNQ_C17." Python is a rich language and there are many ways to code C17 without using those particular functions. The goal should be to get students to solve a problem creatively -- not to follow a particular path.I found the week 4 assignment a real bear -- too long (22 completion sections) in comparison with the others. I'm sure it was difficult for whoever codes the autograder as well. They need to do some more code checking. In UNQ__C8 we were asked to use the pre-coded cosine_similarity function. Initially this was returning the cosine difference, which is quite the opposite. In the middle of last week, after I and others reported the error, it was corrected but this seems to have created a cascade of other errors. For example, in cell UNQ_C9, We are toldExpected Output:[[9 9 9] [1 0 5] [2 0 1]] Which is wrong. A look at the vectors is enough to see that (2,0,1) is closest to (1,0,1) by cosine similarity. I believe the autograder makes the same confusion. I mention this not to criticize our hard-working programmer, who is otherwise doing an excellent job, but so that the errors get fixed ;-) (edited)

A high quality course overall! It helped me understand both theory and the programming mechanics of implementation. The Jupyter notebook guidance was detailed and well-organized!

Enhancement opportunities:

I felt the PCA lectures and PCA function implementation were a bit muddled. Consider illustrating the geometric intuition behind PCA: when 2-D data points are projected onto a line, the "best" line maximizes variance along the line while minimizing the reconstruction error of the data points.

The final notebook assignment is long and contains a large number of function and global variables. It is a lot to digest. Maybe enhance it with a takeaway video that unlocks after the assignment is passed. The video will visually recap what was accomplished by showing the start-to-end pipeline.

The course is a great value for the price!

I learnt a lot on this course - the material about matrices and matrix operations was all totally new to me, so it took a while to get my head around (more background reading links here would've been helpful). Also, with some of the grading cells in the assignments it was difficult to understand why the answer was being marked as incorrect (examples being UNQ_C11, UNQ_C22 in the week 4 assignment).

You must have a very strong knowledge of python to do this course. Concepts are explained well, but work submitted often goes beyond explanation. I found I understood the concepts but had difficulties with the intricacies of pyhon, numpy and various syntax.

The first three weeks were taught really well. But I found the explanations of LSH and Hashtables rushed and they could have given more time to explain it. The assignment in the 4th week was very tough

Ok but coding exercises could have been better structured (e.g. less long functions without easy to run test cases). The exercises could also have been a little more stimulating.

Great course, with interesting examples. However, I am dissapointed with the automated grading system, which wrongly penalized answers and impacted final grade.

Honestly speaking the video materials are not at all sufficient to understand the concepts. I have to watch other YouTube videos to understand concepts.

Very good course from deeplearning.ai team , you will need some background in ML and Python. The support on the coursera forums could be better.

I've got strange feelings about this course. If we are talking about courses from deeplearning.ai, it all started with Tensorflow in Practice specialization.

This course have should have 2 weeks or maybe one, because everything can be accomplished in a few days.

It has very small amount of interesting content.

It is repetitive (some parts of assignments are copied from Deep Learning specialization assignments).

The video lessons are strange - it seems that two instructors recorded same videos and only beginning and ending from Łukasz Kaiser's videos got to the course (which is not fair).

Some procedures don't have description for function's arguments. Some procedures don't have testing function - you don't know if it works or not.

This course should be a part of another course, because it doesn't seem as a complete course. It feels like it doesn't completed, like it is a draft for a full course.

I believe that others courses from this specialization will be better. If you know Russian, I suggest you to take course about NLP on Stepik platform https://stepik.org/course/54098/syllabus. It is much better.

Academically, a step down from deeplearning.ai's previous courses. In terms of technical quality, the media could use improvement, particularly in normalizing audio levels and ensuring a perfect acoustic setup for all lecturers. Coursera should hire some acoustics and motion pictures experts and task them with improving "production values" (look it up).

Shortcomings notwithstanding, this is still a great class and a "must take" for any aspiring NLP expert.

HW check is just awful. I need to use exact these functions (i.e. squeeze instead of ravel).

Moreover, one could complete homeworks without referring to the lectures at all -- zero challenges met.

The last HW is not balanced (11 tasks -- too much compared to the rest).

I would suggest the course makers to redesign the homeworks dramatically.

Thank you first of all. I think the length of videos could be longer for better understanding of concepts. Specifically, I think in Week 4 there are a lot of concepts like ( Locality Sensitive Hashing) that could be better explained.

The assignments lacked clarity. The huge number of variables used for seemingly trivial intermediatory task is really confusing. Also, the assignment documentation and instructions need to be enhanced,

The exercises and assignments seemed to place more emphasis on the coding and less on the theory and study of natural language processing per se, imho.

T

h

e

c

o

n

t

e

n

t

s

a

r

e

a shallow, the lessons doesn't go into details. The teacher only shows what's needed for understanding the week exercises. I won't dare say it's even basic level.