This course was absolutely marvellous! I very much liked Andrew Ng's calm and understandable presentations, the combination of printed and written "slides", all perfectly structured. I feel that the mix of mathematics and implementation has helped my understanding and made me more enthusiastic for each of those aspects than I would have expected me to become. The quality of the exercises was superb!

I started this course by applying financial aid and it's was a good decision to apply for it by writing 2 paragraphs of 300 words altogether along with my personal informations. I believe this course is composed in a very elegant way by the instructors so that students don't get tired of the hard theories. Besides lectures of Mr. Andrew Ng is really pleasurable because he teaches hard topics easily.

Great introdution to neural networks with useful programming assignments (python code) that better clarify what taught in videos. This course begins with logistic regression and then extend the concepts to shallow neural network with very clear notation showing how to implement a neural network from scratch using helper functions. In later weeks what learned so far was extended to deep networks.

The course was really effective as a beginner I have figure out many fundamental ideas regarding deep learning. The video and assignments are really helpful to learn neural networks effectively. And the discourse form is quite good , the metor's quick response and support to figure out mistakes is simply awesome. I will definetly continue my learning journey with you Coursera ..

Thank you Coursera

I would say that this course is a continuation of "Machine Learning" (also mentored by Professor Andrew Ng) but this course demonstrates a slightly different approach to NNs (ML course had only one week of NNs) and goes deeper into deep NNs concept. I would recommend some basic knowledge of Python and some algorithm knowledge from the ML course. Without this knowledge, it might be rather difficult.

This course gives you all basic concepts of deep learning like layers, nodes, activation functions , parameters, cost functions etc.

The programming assignments help you to clear the theory part by Prof. Andrew Ng.

Programming assignments are good ... basic level... clear the picture of neural networks.

Prof. Andrew Ng's teaching style is very nice..

This course helped me a lot .

Thank you very much.

The course is great! You will get a good understanding of the math behind the back-propagation. One thing very important in deep learning is optimizing the code because usually training a neural network can take a lot of time. This course equips you with the skill of vectorizing your code thus avoiding you to write explicit for loops in your code and making your code look easier as well as better.

This is a great course for the introduction of neural networks and deep learning. The materials are very helpful and the jupyter notebook files are straight forward. One minor comment I have for the assignment is it will help if we have an overall picture (or flow chart) of the assignment and all the functions we need to implement at the beginning.

In summary, I would recommend to take this course.

Professor Ng presents neat graphs and flow charts that make difficult concepts accessible. The programming assignments have very detailed instructions and explanations. That greatly aided my understanding of my code and how to write them. I completed his earlier Machine Learning under Stanford University not too long ago so I had a little knowledge of machine learning going into the specialisation.

I think that it's a very useful and detailed tutorial for me EVEN though I had some preliminary knowledge about deep learning. The deduction of forward propagation and especially backward propagation are very helpful in regard to a full understanding of the neural network. I strongly suggest writing something while watching the video. I've benefited a lot from this first course. Thank you Andrew!

Overall great course. Just a couple of points. In the Machine Learning course, there were downloadable course notes and slides on each chapter, would be nice to get something like that for this course too. I am not sure why Python was chosen here and Octave was chosen in Machine Learning. Somewhat liked Octave more, as it was easier to test in that environment.

But nevertheless enjoyed a lot.

Short, up-to-the-point explanations, written notes, lecturing style, interviews section - all is great. And then IPython lab materials, I was really surprised - accurate and helpful, building knowledge and intuition gradually, otherwise it would be very easy to lost the track without this support. The testing blocks introduced in the labs another great idea of this course. 5 stars, no questions.

I have implemented various models previously with the help of TF but was always confused that how these works. Got everything from scratch to implementation. Looking forward to completing the specialization course so that I can add Deep Learning in my CV with 101% confidence. Thank You sir Andrew Ng and all the creators of this awesome course and platform to learn Deep Learning (Neural Networks).

An Amazing class to continue study NN. I took it after the Stanford's ML Class (taught by Andrew Ng as well), and it is just great as a follow-up class. There are a few differences in the approach used in implementation NN in this course and Standford's one, and it was great to learn (and as usual it was very well explained). Thanks to Andrew, from a skeptic I'm slowly becoming a believer in DNN.

This course was very helpful, I knew the theory behind deep learning, but to structure your coding in such an extent is really awesome and using the helper functions. Also it allowed me to revisit the basic concepts of deep learning by Andrew NG which was truly wonderful. Everything complex explained in such a simple way. Looking forward for the next 4 courses in this Deeplearning specialization.

Very well designed course and a great instructor. I've studied deep learning in several other courses for a Masters in Data Science, but the material was never explained as clearly before. For example, I love that animated graphic of Gradient Descent. The detail on derivatives and backward propogation was also excellent -- just the right amount of repetition to allow you to absorb the material.

I enjoyed this course and learned a lot. It was well-structured beginning with starting with logistic regression, moving to shallow neural networks, and finally programming a deep learning neural network. The interviews of well-known researchers were interesting and insightful. The mentor, Paul Mielke, was phenomenal! He helped me find my way when I couldn't find the error in my code.

Excellent teaching, everything taught so well, you don't have to look back again, the course is design in steps so that there is no confusion and Sir Andrew keeps it really simple as well as points out to mistake which are most commonly committed by students. Its a fun course and you will really love it. Though the math is tough but Sir Andrew really makes it simple. Don't hesitate go for it....

Andrew NG is one of the originators of the field, interviews a who's who list in the field of Deep Learning, he's a top-notch teacher, don't expect highly mathematical explanations or extremely complex Python coding, enjoy the insightful comments and clarifications in almost-everyday language, learn Python at an easy pace (for a person with a PhD in Physics and Python / Data Science experience).

Was a great course to start with. NO inbuilt library like tensorflow or scikit, everything was coded by us. So much help was given in coding exercises. This made the exercises easy and interesting. The Week 2 was a bit difficult regarding the order of numpy arrays compared to week 3 and 4. By the end of the course, it was truly encouraging and gave a great insight to L layer deep neural network.

As a linguist with a very shallow mathematical background, even the first few slides with any math in it seemed daunting at first. However, Andrew's explanations are very clear and he covers the same aspects quite a few times. This, together with some research on my own outside of the course material, helped in providing a good, general understanding of how shallow and deep neural networks work.

This course was very useful and taught me the basic intuition for the Deep-learning by teaching the basic architecture of the Neural networks specially Deep Neural Networks.It also taught the Forward and back-word propagation which is the heart and soul of the neural networks.I would really like to thank Andrew Ng for creating this very easy to understand course on Deep-Learning Neural Networks.

Very nice course, I learned practical skills which I can use for my daily work. He could go a bit more into the math details and derive a few things. The practical exercises provided a very big and nice skeleton, so I hardly had to do any debugging to find typos. Most often you get the implementation right in the first attempt. The programming exercises could have been made a bit more difficult.

Very good and informative course. Excellent explanation by Andrew Ng. The programming assignments were very helpful in understanding the concepts especially the hard concepts like backpropagation were very clear.

Just a small request to the course coordinators that if you could reduce the amount of code given to us, it would help. Then it will be more challenging and we can think more.

Thank you.

This is an excellent course to get started quickly on Deep Learning. It covers the right amount of math fundamentals so as not to turn you away which was great. The programming exercises were extremely useful to understand how to apply DL using python. I recommend this course as one of the best online DL courses out there which helps get you up to speed in the super hot DL domain within a month.