The lectures covers lots of SOTA deep learning algorithms and the lectures are well-designed and easy to understand. The programming assignment is really good to enhance the understanding of lectures.
The course is very good and has taught me the all the important concepts required to build a sequence model. The assignments are also very neatly and precisely designed for the real world application.
por Maximiliano B•
In this last module of the specialization, you will learn in details how the recurrent neural networks works. I really enjoyed and had fun with the programing assignments specially the Emojify and the trigger word detection. After the course, you feel comfortable to read all papers mentioned as references throughout the course. Moreover, professor Andrew NG is awesome because he explains the content clearly, it is a pleasure to watch his videos and he provides everything you need to go the extra mile.
por tarun b•
Couldn't be more grateful for having the opportunity to take this specialization. The instructions were just at the right level of illustrating theory in practice, and the programming exercise at the right level to gain intuitions with implementation details. So many rights !!! Personally, I had the confidence that the syllabus is exhaustive and the callouts to research were just great. Overall ... excellent resource I will revisit often. Thank you to everyone who put this together and to Prof Ng.
por Vignesh S•
Thank you, everyone, on the team for such an orchestration of the course. It was excellent to get to know the concepts of deep learning and it increased my interest in the field exponentially. A special thanks to Dr. Andrew NG for those explanations given in detail. This course was really interesting and it definitely overturned my attitude towards NLP as at first, I thought this is gonna be a difficult field of AI.
PS: Keep that ever-smiling face of yours the same Andrew Sir. Thanks a lot.
por David T•
Very effective mix of theory and practical examples, cemented by practice exercises in form of the programming assignments. The guided instructions of the programming assignments are as valuable as the lectures. I would recommend putting the "errata" readings *before* the videos containing the errors (as was done in an earlier course). I also notice that some quiz questions pick up on nuances that were quite underplayed in the lectures, but going back over my notes, did find them.
por Amey N•
Smooth and hands-on walkthrough of basics of NLP and speech recognition. The flow of the course is very well-designed.
After having completed this specialization I can confidently say that I have a much better understanding of Deep Learning than what I had before I underwent the specialization. This includes the depth and breadth of DL, various models, their challenges, advantages & disadvantages, end-to-end pipelines, optimization techniques, background calculus & math, et cetera...
por Brian H•
Amazing course overall. Prof Ng's diagrams are the clearest explanations of DL models I have found anywhere and it's that clear a ton of thought went into planning the notations. The assignments are exciting and surprisingly fun. One could say that there is a little too much handholding throughout the assignments, but I understand that this course is more about the heuristics. Again, it's fantastic course overall and the resources provided throughout are truly unique to Coursera!
por Kuntal C•
This was my first AI course and I really made significant progress in my understanding of foundations of deep learning with this. Thanks to Professor Andrew's very informative course videos, grasping the complex concepts became possible. The quizzes and the assignments were challenging, made possible for me to use logic and develop new coding skills to go at it. I would recommend this course to everyone interested in AI/ML. Thanks to Professor Andrew for making this course.
This is an amazing course, it Provides a great Help...i have learned lots n lots of stuff about NLP, Learn about recurrent neural networks that work extremely well on temporal data, word vector representations and embedding layers --that are explained in a concise manner, and more importantly I love the Attention mechanism, the model that understand where it should focus...... its attention given a sequence of inputs.... amazing amazing ..highly Recommended.... Thankyou
por Alejandro R•
This course was a great introduction to the world of RNNs. Starting from basic sequence models all the way through RNNs constructed with Convolutional layers, LTSM layers, GRU layers and wrapping up with the Attention Algorithm. It is great base work to start a Deep Learning career. The course is very well structured and the resources in the forums were always life-saving. Very grateful for this course and I am waiting for the Advanced Specialization from Deeplearning.ai
por Janith G•
Really good course for RNNs with NLP. Recommended to anyone who has completed the first four courses of the specialization. A thing to notice is that the last programming assignment is really hard to save and submit to your servers though it was pretty well organized.
Also I would like to thanks Coursera and Prof. Andrew for bringing ML DL and AI to a level that a student can understand without any useless long mathematical proofs. Thank you for giving this opportunity.
por Artem D•
I really liked the whole Specialization, it is great: clear and interesting!
But the last course seemed very difficult to me: may be I've been pretty overhelmed (I've completed the spec in less then in a month), may the topics are much harder then in previous course, may be Andrew Ng wanted to cover too much items in short time. It seemed to me hat CV course was more clear.
Nevertheless I rate this course @5 stars and beleive that the spec is PERFECT!
THANK YOU, ANDREW!
por Carlos V•
Another Excellent Course from Professor Andrew Ng. The detail in the explanations are excellent, and the provided exercises using Jupyter are super fun to complete and put to the test your knowledge offering you at the same time a library of ideas and models to use in your future projects. I enjoyed this last course in the specialization quite a lot, thanks very much to Andrew Ng and the Staff from Coursera. I hope to see more courses like this in the future.
por Rodolfo V d A•
I love this course. Maybe about 1 and half year I was trying to learn DL. I thought about giving up because I wasn't able to learn, so, in May of this year 2020, Coursera opened its courses to undergraduates. I do not thought twice, at the same hour that I knew that news about free courses, I began this. Then, after that time I want to thank to Professor Ng and all colaborators on that site. One day I will be one of student of Professor Andrew Ng in Stanford.
por Prithvi J•
A greatly knowledgeable course! I learned a lot about Natural Language Processing and explored RNNs, LSTMs, Word Embeddings, Seq2Seq Models, Attention Mechanism, etc. The course focuses more on the concepts along with providing the essential math. It was fun to implement Language Models, Neural Machine Translation & Speech Recognition. I would surely recommend this course to the ones who are diving into the world of NLP, and need a perfect introduction to it.
por Huanglei P•
This end course is a little more complicated than the previous ones, especially in programming homework. However, it also inherits the merits of the special, gives learners the basic framework of sequence models. What impresses me most is the lesson of "Debiasing word embeddings", it shows that AI could be designed to do more against human stale thoughts, which sets up a good principle for designing AI. Yes, it should be taught to new learners of AI.
por Andrés G D•
Finally... Every piece of effort was worth it! After so many hours, now I understand how proud we can fell of completing these amazing courses! The best one I have tried so far, definitely made a difference in my professional views but above all, it confirmed my expectations: this is the activity sector where I want to develop, the work in which I want to grow without any doubt.
Thanks Andrew. Thanks Team. Thanks to everyone who made this possible.
por Marcus H•
This regards all 5 courses of the DL specialisation.
1st of all: great work, it gives a much broader perspectives.
Room for improvement: sometimes the assignments become much of a "Python riddle" where one has to fiddle a lot with language technicalities and loses time for actually playing with the DL subject
2nd: please improve the submitting and savin g behaviour of the notebooks in the new LAB system. It is really painfully slow and unstable.
por ANSHUMAN S•
This was the most difficult and most interesting course i had in all of the five deeplearning.ai courses
but after doing all the 7 assignments i feel like i learned a lot and encountered with some of the amazing thing which i wondered how they are done . Once again I thanks to Andrew Sir and other teachers for beautiful lectures and perfect quizzes assigments and at last a heartly congrats to Coursera for giving this platform to me.
por Mihai L•
Will give this course also 5 stars. The assignments were easy but required some knowledge of Keras. So you have to invest some time on their site.Otherwise it's like fitting pieces in a bigger puzzle. Most pieces are already layed out for you .. you need to just fit your small ones.
I realize though that deep learning requires a lot of practice and experimentation and completing this course (and specialization) is just a tiny first step ..
por P S R•
Course contents and coverage was best. Duration of 3 weeks is little too short to really understand all the details of programming exercises. May be extend this to 4 to 5 weeks and spend little more time on speech recognition, music generation and other audio data processing would have helped.
Unlike all other earlier modules, this one had many issues with grader and many errors in note book templates. Hope these will be addressed in future.
por James B•
Wonderful course, expert instruction from Prof. Ng. I can't recommend the Specialization enough.
The choices of architecture and of hyperparameters for the assignments' network could have used further explication. Another desire left unfulfilled was that I would want the sequence models course doubled in all dimensions, ie lectures, assignments, etc. It was all over too quickly with questions lingering. Further study required!
por Weinan L•
RNN, LSTM, GRU... fun stuff even you don't focus on NLP. As always, Andrew makes complicated things simpler. I certainly will keep all the course materials for future reference.
It may be easier to follow other online course, but this course will teach you not just how, but also why...
Read coding instructions carefully and pay attention to details, otherwise you may end up with hours of debugging. That's what happened on me, LOL.
por Virginia A•
Sequence Models are a though subject. many people, during working meeting, mention them as the final resource and solution to everything. I feel I better understand the nuances of them thanks to this course.
I personally enjoyed some of the extra reading ( original papers quoted at the bottom of the videos). Sometimes is hard to navigate in the large sea of publications. It is nice to be pointed towards some piece of reference
por Chris D•
I go back and forth on whether the time-saving aspects of the Python Notebooks are worth the reduction in ML coding experience. I suppose these aren't coding classes, but I also feel some of the concepts aren't cemented as well as if the students were led through a more challenging, trial-and-error experience. That's hard to do, though.
Overall, I recommend the specialization. Maybe just be sure to play around offline, too. :)
This series of course provides a comprehensive overview of NLP algorithm and different applications. I really enjoy the projects the deal with audio files. The course skip the linear algebra and differentiation part that not everyone wants to look into. But I hope it will be better if we could also implement the data processing functions of different types of sequential inputs, since data preprocessing is also significant