good

.

I'm a master/graduate student who took an NLP course in Uni.

I think that overall this is a very a good introduction to the topic. Some concepts are really well explained - in a simple manner and with a lot of jupyter-lab code to experiment with.

In general in this specialization - the first 3 courses are good. There are some quirks (e.g. why Lukas is needed at all? He doesn't really teaches, just passes it on to Younes) but nevertheless I learned from it. And I think they have good value in them.

The 4th one, however, is completely disappointing. First 2 "weeks" are confusing, not really well explained, but somewhat "bearable". The last 2 weeks are complete sham. They claim to teach "BERT" and "T5" but don't really give any value. You're better off going elsewhere to learn these concepts.

If it wasn't for this, I would give the overall experience a 5 stars, but because of this, I think the overall is more like 3 or 4.

The courses have really enlightened me on NLP. I had no idea about the techniques. I'll give it 4 stars, because the course instructors have a monotonicity of lecturing as if reading from a teleprompter with a fake synthetic voice. It sometimes gives me a headache and I end up muting the videos and just reading the subtitles (which a lot of times don't make sense and are short paces so I have to freeze the screen, and open 2 other windows and read the lower caption text). I have been doing many courses on this platforms, and even though the instructors are on the top of their fields, but the way they deliver the courses is just "sometimes" and "not always" painful. I am sure this is not how they teach there own classes, especially in Stanford. Even though the course is 50$ per month, a think it won't cost the instructors much to show some authentic enthusiasm.

Es bastante completo, y en general, claro; salvo un detalle: explica demasiado superficialmente PCA, pero luego para la tarea hay que implementarlo manualmente. También tiene algunos bugs desde hace meses a pesar de haber sido reportados con solución. Además, las lecturas posteriores a los videos a veces son escuetas y le hacen falta algunos diagramas que se ven en el video (conviene sacar capturas de los videos para tomar nota).

It is quite complete, and generally speaking very clear, except PCA: it's covered only superficially but it is required to implement by hand on the assignment. Also it has some unsolved bugs since several months ago, despite they were reported with solutions. Also, the readings after the videos are sometimes narrow and lack of some diagrams shown on the videos (it is useful to take screenshots to take notes).

The content of the course was really interesting an engaging. But the assignments mostly only helped in understanding the details of the algorithms and processes. It would have been nice to get to learn how to use state of the art libraries, which would've been more practical. Although, in fairness, anybody who completes this course should be able to make use of off-the-shelf libraries. Another point was that when the instructor was narrating the slides, his intonation was occasionally a bit off, making me lose track of the subject and having to re-listen few times.

Personally speaking, this course is great. I have a background in Math so it would be very easy for me to infer all the mathematical details from a general thought or frame of the method, but I can imagine that people with other backgrounds may suffer from a lack in rigorous proofs in this course. Nevertheless, the course does provide new thoughts for me and make me familiar with some practically useful tricks in NLP. I would recommend this course if you are working in a field other than NLP and want to learn something about it or if you are a beginner to NLP.

Good and instructive course. Minor problem of this course is that authors tried to make it less intimidating for students that lack math skills, and that results in that some explanations are not rigorous. For example the principle of extraction of word embeddings from CBOW model is explained purely on the basis of dimensionality of weight matrices whereas it is leaves a lot of questions unless one studies supplementary materials.

Good for the basics of NLP. Good mix of examples from classical NLP (e.g. n-grams) and neural nets (e.g. embeddings). As usual from deeplearning.ai, great motivating examples such as autocorrect and autocomplete to help us understand the materials. The neural net examples could do with more equations as in other deeplearning.ai courses.

The quiz for week 1 contains topics from week 4. It has not been changed in 2 weeks that I was on this course. Except for this lacuna everything else was very nice. The course is well structured and the assignments made me to think and revise the course material thoroughly. In a nutshell this was an excellent course.

Thank you for the Lecture. I enjoyed it a lot! One thing I did not like too much was reading aloud and fast complex equations. I got distracted a lot when that happened. Also the Grade of the programming assignment is very slow and some additional verification of the programming task would be helpful.

There is a confusion in week 1 practical quiz. It seems that this task from week 4. There are conusions in contests but in general this is good course, because we come through the program from simple to difficult tasks and make necessary computings and functions from libriaries from scratch.

A good course overall. I wish the assignments were a bit more challenging though. Still, we have covered a lot of ground. And for those who know nothing about the word embeddings, I think this would be a perfect first course to take. So all in all, time well spent.

I found the course really helped to reinforce my understanding about importants concepts like n-grams, HMMs and word embeddings. The labs are pretty well spread out, and by the time you get to the week-ending assignments, you have all the info you need to complete.

It will benefit if we can apply the knowledge at work while learning. Fourmulas in videos, readings and assignments are sometimes different. There is almost no help in community forum if we are stuck with assignmernts. It helps if we get help from TAs.

I love this course, If you follow along with book An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition by Daniel Jurafsky and James H. Martin will help you to understand the core concept and mathematics

The lectures on the Viterbi algorithm were a little wooden and there were no summary text (reading) tasks (as there often is in other deeplearning.ai courses), however this is a worthwile and informative course.

In the first and second week the exercices have some unecessery pranks in the data formatation just to make the exercice harded, but it take out the attention for what matter in the course that is NLP

Very good to see how the "from scratch" concepts are presented; nevertheless, I have the feeling that a very little "real use case" problem has been presented, with tiny sentences being analyzed.

Week 4 Lab Assignment could be made a little bit tougher. The backpropagation derivation of W1, W2, b1 and b2 could have an optional reading for the interested reader. Otherwise, amazing course!

Good course, but the lecture notes in week 2 can be much more improved. Understanding Viterbi algorithm without visuals and animations was very difficult. Apart from that, great course!

The course is exceptional in its own way by bringing people to the understanding of probabilistic models. Crisp & Clear. But one need to explore & practise more to gain expertise.

I felt like I learned some new things from this course. Some of the maths was not as rigorous as it might have been. For example, the proof for Levenstein wasn't complete.

I liked to learn about Word2vec in the week 4, using Continuous Bags of Words, step by step. It helped me a lot to understand how to transform words in numeric vectors.

The Viterbi algorithm introduction is a bit hard for us to follow. Probably some writings may help guiding through each steps.