If you have already tackled some machine learning problems yourself and really want to understand the inner workings of deep neural networks, this course is for you ! After this course, you will be able to program basic deep neural networks even if the lessons on hyper parameters tuning are left for the next courses.

Really quite an experience to go through this complete new knowledge space. The course is well structured and very practical for any professional to retool itself with the latest AI development, The lecturer - Andrew Ng and the DeepLearning.AI community are extremely helpful and encouraging. Many thanks to all.

A great course which is designed carefully to learn as much as possible in a simple manner. We get a good perspective towards deep learning/AI after this course. Also, it's a great foundation to build further I suppose. Because, I already got an idea after this course about how to move further with deep learning/AI.

The only thing I miss, is to be able to debug programming assignments. I googled, that there is generally an option to debug jupyter notebooks, but not not figured out if it works with Coursera yet. It would be great to have this option - it would make an already wonderful learning experience absolutely perfect. ;-)

Throughout this course, I learned the foundations of deep learning, with a comprehensive understanding of neural networks and key parameters in a neural network's architecture. And I got both an intuition and a deep understanding (I mean, familiar with the math thesis) about forward propagation and back propagation.

The course is great, just it needs some more videos explaining what we did in assignments, although it is clearly, but videos helps :)

To the new learners out there, go for it , although you MAY need some slight knowledge of Ml things. Just check on youtbe or somewhere and get some basic idea and then ready to start.

Excellent expiation of deep neural network architecture with realistic examples. Nice math explanation behind neural network like forward propagation, backward propagation, loss and cost function. Assignments are really helpful to apply theoretical knowledge. It helps me to get much confident on Deep Neural Network.

A very good introduction to deep learning where I was able to get the proper intuition and understand the minute working of the forward and backward propagation. The course also helped me to understand the mathematical derivations for the backward propagation specially and how it is working for the gradient descent.

Andrew Ng explains things in calibrated steps, without skipping the math, all the while ensuring you get the big "intuitive ideas" along the way.

The interviews with AI Deep Learning "heros" are fabulous because they inspire and provide context for how the field is evolving.

The course is challenging but not daunting.

Though I struggled initially with understanding the math (been over 30 years since I took these classes in secondary school), a little remedial training on you tube got me back to a reasonable understanding. I appreciate the education on Python and now I have a much better understanding how to train neural networks.

It was an awesome course and awesome journey as well. I am now much clear about the concepts of building a neural network. I believe I will once again work upon the assignments to brush up my skills. Thank you so much to the whole team for the efforts that you put to make this course a success.

Thank you so much sir.

Excellent intro course on neural networks. Even though it is best to have completed the machine learning course prior to this one, the content is not much more difficult. The program assignments are particularly well made and touch upon every aspect taught in the course, they can almost be used as the course itself.

A great bottom-up overview of deep learning concepts with a decent amount of theory and a good emphasis on practical applications. I like the fact that you get to build everything almost from scratch to learn the building blocks (rather than starting with a framework) and then build it up to a practical application.

Do deep learning with just Numpy. THAT is by far the best thing about this course. No barrage of libraries being introduced and the additional jargon and implementation idiosyncracies that come with them. Pure matrices. You get a lot more out of it if you program all the exercises from scratch and I mean A LOT more.

Andrew Ng at his best: perfectly clear, attention to detail, highlighting potential gotchas. Practical training with hands-on implementation of Neural Networks in Python/NumPy. In my opinion, it's a better choice than Matlab/Octave. As a bonus, learnt some useful tricks relevant to other NumPy applications. Great.

Great course. The clear, step-by-step progression from Logistic regression to deep learning pretty much completely de-mystifies deep learning. Thank you Prof. Ng. There are a couple of surprising little errors in the example code here and there, but that just ends up improving the learning experience, as it happens.

Enjoyed thoroughly. I think a bit harder (and more realistic) assignment would have been better. I was unhappy with 80% accuracy in cat vs non-cat pictures. At the end of final assignment my program classified the image of a watch as "cat". May be I am jumping ahead and there will be better models in future courses.

A great introduction to Neural Networks as expected from Andrew Ng. The Jupyter notebooks are very well documented. Even though I was familiar with most concepts explained in this course and am more interested in the other courses of the specialization it was well worth doing this refresher. Looking forward to that.

The course was challenging. Right from the start Andrew pushes you to think about neural networks from a mathematical point of view, which in turn makes it easier to appreciate NNs. The pace of the videos is perfect, concise enough not to be boring, but long enough to be able to explain the intricacies of the topic.

First of all I would like to thank coursera to gave me opportunity to follow this great course. It is very much valuable course for those who starting the data science field. I must admit this will be more than 5 star course and recommend for my close friends those who studying in data science and machine learning.

Very good course content, learning the concepts and implementing them has made learning much more fun. One thing is that since I had an earlier knowledge about gradient descent, I could understand but a complete newbie will face some difficulties. So an optional video explaining Gradient Descent will be very useful

Course is very ideal for the people with linear algebra background .

I see many times Professor Ng saying you don't need to worry about complex maths etc.. but he spends most of the time explaining the equations.

It would have been more beneficial if the real world problems and solutions are discussed in the course.

Excellent course. I took this course after completing the Coursera Machine Learning course by Andrew Ng. Nice to have both Matlab and Python experiences. I really liked hearing from the Deep Learning experts in the included "heroes" videos. Andrew's teaching style is the best I have come across. Highly recommended.

The way mathematics is explained is very intuitive. If you know basic calculus and matrix multiplication, you will easily grasp all the concepts taught in the course. Moreover, by the end of the course we implement our own neural network without using any library like TensorFlow which really adds to the confidence.

Andrew presented the material in a way that was very easy to understand and the exercises were very helpful for learning how to code deep learning models. The format of the course is exceptional with videos and text of the lectures to go back and review any information that isn't understood the first time through.