This course was very straightforward to understand. The lecturer taught it at a level that you didn't drown in the math, but if you were interested in going deeper in the math you could. The programming exercises were straightforward to complete yet I learned a lot and gained confidence in the material that was taught. I would highly recommend this course for anyone who has some programming skills but does not have skills in Neural Networks or Deep Learning.

One of the best introductory courses, taking the learner from simple logistic regression, through computational graphs and teaching the core of neural networks viz. the back propagation algorithm. The glide from zero to some solid foundation on neural networks is something I thoroughly enjoyed as learner. A probing quiz and programming assignments that pushes the learner to seek finer aspects of the neural networks is something that is missing in this course

Having done multiple Udacity Nano Degrees and other deep-learning/AI courses on Coursera/edX as well, I can say that this is probably the best and most detailed course to master Neural Network mechanics esp. back prop. Videos do a great job in explaining complex and confusing concepts in easy to understand style. Assignments cement the understanding further.

Kudos to the Prof Andres Ng and rest of the deeplearning.ai team for putting up such a great content.

I think this course is thought for people who are beginning to explore the world of IA, and people who don't have much programming experience, probably that the reason of the excessive help in the assigments, if you have studied before machine and deep learning and you have already an idea of what is it, you'l probably find this course easy and tedious, but in general I think this is a good course for begginers and I feel I learned a lot and, I'm satisfied.

This is simplest, most concise explanation of back-prop I have ever seen. This was my first Andrew Ng course, but this is in line with his reputation.

This was taught at the right level with respect to the complexity of the material. Andrew gives a mathematical basis for everything, and the assignments give you an intuitive understanding for how to organize a neural network project. At the same time, the course remains practical/grounded. That's hard to do.

The course is magical. I suspected I would have several road blocks in completing it. But the simplistic approach by the instructor and well-guided exercises along with coverage of prerequisite basic knowledge wherever needed made it a smooth undertaking. The course achieves its promise and the once unknown field of Neural Networks and Deep Learning is now a familiar topic of interest and has set my path towards working on more advanced topics in the field.

Good introduction that focuses on developing a good understanding for how neural networks operate. The level is not too deep, although the optional videos provide additional detail.

The level of python programming required to complete the course is very low as you get lots of help with what packages and functions to use.

Very enjoyable and who doesn't need a cat picture classifyer!

Looking forward to the more advanced classes within the same speciallisation.

As usual with Andrew's courses the assignments are really helpful. The only thing I think that needs to be improved is the last section doing backpropagation over multiple layers. There were some helper functions that we called from another file but I felt we should have seen those more easily to learn them. Also was a little confused about the indexing of some of the stored parameters. Would like to have seen the equations next to the code. Awesome course!

A great introductory course into deep learning. I highly recommend if you're looking to secure a strong foundation before moving on to more difficult concepts. Regarding prerequisites, having basic skills in Python (for loops, syntax, data types) is sufficient to be successful in this course. While a background in calculus and linear algebra may be helpful, it is not required as the instructor will review the concepts needed to get started. Happy Learning!

This course is very suitable for learners who want to grasp the core ability of NNs. You will learn some fundamental knowledges as the base of mathematics firstly. Then it will step-by-step to teach you how to implement the power of Python to achieve linear-Regression, logistic regression and NNs. Especially the part of forward and backward propagation. Thank you Coursera, you give me the opportunity to learn some top knowledges to change my life for free.

The practical approach to working thru the mechanics of forward and backward propagation is very valuable (as i'm sure Andrew Ng knows). It was helpful to get a solid footing in how to construct the models and execute the vectorized functions to expedite the computation of the parameters in a model. As a math nerd i was hoping to see a little more in the chain rule derivations but that definitely isn't critical to the successful completion of the course.

I'd read a lot about the theory behind neural networks and the back propagation algorithm was a topic that I very often had to keep reminding myself of whenever I returned to neural networks after studying something else for a significant period of time. This course allowed me to see back propagation in a different light which I believe will make its retention easier. (activations -> forward, gradients ->backwards then parameter updates with the gradients)

As expected professor Andrew Ng (and the team who was involved in creation of this course) made an awesome course. I completed the course from Andrew Ng for Octave before this one. And still this course helped me to kick start doing ML on Python, and helped to systematize and deepen my knowledge. I am jumping into the second course of specialization right now. These courses are great product, and I would recommend them to everybody who want to learn ML/DL.

This is the first course that I attend on Coursera. I am really satisfied by this platform.

This course was very interesting, and thanks to the experience and ability of teaching of Prof. Andrew Ng, I could learn many new things.

Fortunately, the course was not only focused on theory, but it included also some training with Python, so it was also a good opportunity to learn a bit this programming language, which is new for me.

Thank you a lot Prof. Andrew Ng.

I had a great learning experience with the "Neural networks and deep learning course", Professor Andrew Ng and his team have done an amazing job at breaking the complex theories into simple explanations.

This is not a course for beginners because some Machine learning and programming skills with python libraries such as numpy and scikit-sklearn is a prerequisite, if one does not have the prerequisites it will slow you down in the programming assignments.

Yes, as I always believe it is NumPy that should be mastered before diving into DL frameworks. Grreat introductory course. The derivation of the equations of forward-prop & backward-prop was important. Programming exercises were so useful to practice the basics one more time and more importantly, to visualize what's happening inside in a shallow and deeper network.

Highly recommended course for anyone break into AI/DL or just interested to learn ML stuffs.

I enjoyed the course. My issue was that my Jupyter notebook would often loose connection to the network. Maybe that was b/c I left it open too long; but regardless when I tried to reconnect the kernel it never really worked and I had to launch a new Jupyter session.

It would also be nice if:

(1) it was easy to see a table of variables (like Matlab whos command)

(2) it was easy to open up some of the subroutines that I'd already written for reference

-Kris.

I have been working on using classifiers for quite a long time already. But in the last years I had the feeling I have not kept up with the recent developments and that I am still lacking basics. This course gave a very good refreshment for both of my concerns. The lectures are well-understandable keeping a certain intellectual level, though. The same counts for the programming excercises. Doing those with the Jupiter Notebook environment was really fun.

A well designed course. Good for giving you a good starting point. The intuition behind matrix multiplication and how to think about the correct dimensions was great. It was also notable how well structured the helper functions were which made the later implementation of the neural network almost a no brainer. The most crucial outcome of this course is that it will give you a logical, sensible and steady way to approach a problem rather than going nuts.

Nice explanations and complimentary material. Ideal for people who has already a slight knowledge on how neural networks work and want to review and assimilate concepts in a structured course.

If you are not familiar with linear algebra or calculus you can get through it, but I would recommend an introductory course on both subjects so you will be able to better understand the details of the mathematical formalism.

I will keep on going with the next ones.

Intuition is the keyword here for me. I didn't feel the need to memorize anything. The content is designed in a smooth way to raise your intuitive understanding. In some cases, I needed to re-watch a video a couple more times but with every time there is an added value. Now that I have completed the course, I plan to re-watch all the videos again and review the assignments and the quiz and I am almost certain that it will not be a waste of time to do so

I really enjoy this course. It was designed in such a way that if you follow the viedo/lecture carefully, the problem set or quiz will naturally be solved with the minimum amount of effort in programming. It is really friendly for beginners in Deep NN, who have intermediate programming skills but want to soak fully into the knowledge of Deep Learning. I really appreciate Andrew's fabulous teaching skills that give me a quick jump-start on this subject.

great course, clear instruction. much better than Machine learning course taught by Dr. Ng. (Machine learning is already a great great course, so this course is GREAT GREAT GREAT course). Now, I am crystal clear of neural network and backpropagation algorithm. Although in real work, we may simply use a commercial software to implement, understanding what is happening under the hood is really important. Thanks Dr. Ng and can't wait to start next course

Very excellent class, learned a lot about Neural Networks' concept and practices. Andrew Ng can really turn complex concepts into easy understanding teaching. I see a lot of people complained that the programming assignments are too easy, but I feel that is just not this class's focus. There are some other courses focused on python programming. The value of this class is to let you understand the complex deep learning concepts and how to implement it.

It is been a great learning course.

Since I do not have computer science/engineering background, it was a bit difficult for me to grasp the Object-oriented programming usage and Forward recursion used in week 3 and week 4 programming exercises. But still, I was able to complete the exercises and understand the deep -L-layer network, thanks to such an amazing guide provided in the exercises.

Man! I love the way Andrew Ng explains the topics. I am a Fan!!