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.
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.
by Kostas H•
The best online course I've seen anywhere about recurrent neural networks! Prof. Andrew Ng explains everything in such a simple manner. For example, understanding the structure of LSTMs is quite challenging, but Prof. Andrew Ng explains it in a very easy to understand fashion. Likewise with GRUs, Seq2Seq models, bidirectional RNNs, etc. And the code exercises have very beautiful and detailed explanations.
by Guruprasad S•
Thanks Professor Andrew Ng and team for the deep learning specialization. The course material was well designed for online learning. The assignments were perfectly manageable with a few hours of investment every week and the learning was very effective. Last but not least, I found Professor Ng's wisdom, insights, tips to be invaluable to anyone regardless of their level of expertise in machine learning.
by Shishir M•
This was the best course among 5 course specialization. It was well designed, structured and application oriented. Assignments were pretty fun to solve as they involved solving real world problems. This course gave me direct exposure to industry level problems and helped me gain more insights towards the future of deep learning. Because of this I am really excited to continue working in deep learning.
by Rohit K•
Hello Andrew, I am a big fan of you. Learning from your every course. Very unfortunate that I can do that remotely only.
One thing that I want to mention - Can we have lecture notes on coursera, just like the way used to in CS229 that we can read before coming to next lecture. I found that that was very useful in understanding when things get harder.
Thanks hope we can improve coursera in that matter.
by Daniel C•
The Sequence Models course covers state-of-the-art deep learning methodology as of 2018. The instructor is awesome. The assignments help solidify concepts presented in lecture videos. One nitpicking comment. This course, being relatively new, was less polished compared to the other courses in Deep Learning Specialization. I'm sure future updates will eliminate glitches and errors in the near future.
by Raimond L•
Good course to help you understand how sequence models work and how to apply them for various problems. Majority of topics are explained quite well. Practical problems sometimes could be a challenge, but every problem has hints and a bit of theory provided. Overall this course was a very positive experience and I do recommend it. Special thanks for the people who made this course possible.
by Dmitry T•
I liked that this course was a bit harder than others in the specialization (well partly because It felt like notebooks were made in a bit of hurry here) but it was a good thing for me, since I had to think more on the programming excercises, read Keras documentation, derive backprop equations - and I believe such engagement with the topic really allows to understand and remember it better.
by Mary A B•
It's been so rewarding to apply what I've learned in the previous courses of the Deep Learning specialization to time-based problems. I feel I have a better understanding of how some of the "magic" technology like virtual assistants and speech recognition work. While the material in the first four parts was also very useful, the specialization would have felt incomplete without this course.
by Honza Z•
Where shall I start...? This module was by far the hardest I made, but I'm really glad I was able to finish it somehow (Searching of my own typos was quite challenging task and I thought my head explodes). Anyway this set of courses is great and I will continue further in my path exploring the world of AI. Thank you guys for the effort you spent to share your knowledge with us. Great job!
by Leandro O B•
Another outstanding course about Deep Learning.
It teaches Recurrent Neural Networks from the basics up to industry applications such as Speech Recognition and Natural Language Processing. The programming assignments are extremely useful to build strong understanding of the algorithms, which we code "from scratch" with NumPy before using higher level frameworks such as TensorFlow and Keras.
by Abe E•
It's a great class, and Andrew Ng is a great instructor. I wish the exercises were a bit harder. Since the course is aimed at all and I am coming from a graduate degree in the sciences, I realize it's hard to cater to all educational backgrounds. I would have liked to see optional/honors exercises to get us more involved. Other than that, I loved the class. Thanks so much for teaching it.
by Patricio G•
Comencé esta especialización sin conocimientos de deeplearning en absoluto, hoy habiendo finalizado la especialización tengo una basta noción de este mundo tan apasionante. Quiero destacar la facilidad con la que Andrew transmite su conocimiento, es un instructor de otro mundo!. Feliz de haber realizado la especialización y de continuar por este camino. Gracias a Andrew Ng. y a Coursera.
by Congyuan Y•
This is an incredibly great course for learning Deep Learning. The course lecture videos and the programming exercises are both so well designed! By learning this course, I have got a comprehensive understanding of Deep Learning framework, as well as the hands-on experience of using deep learning to solve real-world applications. Thank you for providing this wonderful series of courses.
by Simon R•
Loved the course, Andrew is a great teacher; very impressive ability to explain and give intuition. I can really see how I can build upon this course to help me in what I am doing at work. I think there is definitely some room to go deeper on some of the topics e.g. don't just teach sequence to sequence but also broader uses of recurrent networks. Maybe a follow-up course? ... please???
by Sampath T•
First of all I would like to thank all of the people given me opportunity to follow this course. This is the toughest course I followed so far with latest greatest technology stack in 21st century. Over the past few months I gained a lot of knowledge and experience from all of the courses and I hope now I can apply the knowledge of deep learning specialization for my future projects.
by Fabrice L•
This module of the specialization is a bit more complicated than the others; at least to me, I found the concepts more difficult to grab.
Anyway, thank to Andrew and his team for this amazing specialization. The lectures are great, the assignments are fun and have interesting examples. A huge amount of knowledge along all the courses. You can tell there is a lot of work behind it.
by Apoorv V•
I was about to give this course a 3-star rating unlike the other courses in the specialization, which I have rated 5 stars. The reason for that was the programming exercises in week 1. They are not interesting and do not impart a lot of learning. Please consider improving those. The reason I still gave 5 stars is because of the amazing programming exercises in weeks 2 & 3. Thank you.
by Aman D•
It was really wonderful course. From here i had learn most important algorithms like LSTM and GRU. I had learnt how to deal with sequence model. Hope I will be better perform on different data.I am really really thankful to coursera to provide such legend teacher. Once again thanks for this beautiful course. Now i had clear my goal. I want to be something in the field of ML and Ai
by JC Q•
In the continuity of the 4 previous modules, the Sequence Models course is of very high quality, the material is concise but cover a wide range of applications and methods, and is delivered with consistent clarity. The programming assignment gives very good hands challenges. I highly recommend this course to anyone interested in natural language processing or speech recognition.
by Mashrur M•
This is the last course of the Deep Learning journey, and I felt like a learned a lot in it. Sequence Modelling is a different beast compared to non-time series models, but I've mastered it thanks to this course nevertheless. I would recommend this particular course to anyone who has a moderate understanding of deep learning and wants to get into time series analysis and nlp.
by Peter V•
A succinct overview of a number of ideas in sequence models. Some of these were covered in an NYU course I took 4 years ago (embeddings, LSTM), others I had heard about but hadn't had a chance to look into (attention). The assignments were set up to be pretty easy, but I think trying to do them from scratch rather than by filling in code would make for a pretty good project.
by C L•
It is a well designed course (especially for week 1 - week 2) which gave a comprehensive view of the sequence model. However, week 5 material can be better and clearer. To be specific, additional hints can be given in the coding exercise. Video can be better aligned with the material in the coding exercise. Thank you Andrew and all the contributors for this amazing course!
by Guy M•
Great introduction to sequence models/RNNs. The real-world examples were very illuminating. Again, as with the previous course in the specialization, I felt some details of how to run/predict NNs using keras were lacking, which could leave a student floundering if they've never used keras before. This is in contrast to some other, much easier, tasks where hints were given.
by Alexander G•
Out of 5 courses offered I think this was the most exciting one as it combines everything learnt so far and teaches how to combine different NN modeling techniques to achieve desired classification/prediction features . For example, it is pretty much clear how one would go about building an app that would describe a picture to blind people and do that in many languages.
by Kévin S•
This short 3 weeks courses will make you work a little, exercice take at least twice the time write. You will learn about the famous LSTM, and how to use it on various tasks.
I'm not sure the 'translation' tasks is a good example but there is lot about it. Not a good example, because it is not state of art, and in the 'translation' business there place only for the best.