I had an incredible journey through the initial concepts of deep learning, and I cannot believe that I have just built a L-Layer Neural Network capable of recognizing pictures of cats, even the most strange ones I could find...

A big thank you to Andrew Ng who is very clear, pedagogical and didactic, and manages to convey in relatively simple terms what appears today like extremely complex notions. Well done and thank you.

I am someone who has a background in some basic machine learning and intermediate Python programming skills. However, I have always struggled with rigorous mathematical implementations that are required in deep learning. Dr. Ng has been the only person who was able to sufficiently explain for me. This is a great course, and I highly recommend it to anyone wanting to learn more about the inner workings of a Neural Network!

Great course! I've used neural networks before but it is really nice to understand the detailed process of it. If you have strong mathematical background you will really enjoy this course but you can also enjoy it without trying to understand all of it. It would be nice if some programing assignments were more open. The current ones are too guided for my taste. However I understand this facilitates grading the exercises.

highly interesting. I loved the course approach that aims to teach not only few lines of code but also the why/how/when those lines became important. ML Evolution and interviews to field heroes is something powerful I could find exclusively here. On the technical side the clear explanation, and related examples, of vectorization and all the steps, layer by layer, index by index, of a fwd and bkw propagation is priceless.

This whole specialization/set is a great foundation on the theory for deep learning. It provides the why rather than just teaching various tool/functions. Andrew Ng is a very good, positive, encouraging, and respectful teacher (especially to one trying to learn the topics). Online learning feels so much better than grad school one-time pass on classes because I can go back to the sessions and refresh my understanding.

I won't say the course materials are brilliant, nothing else can be expected from the greatest machine learning teacher Andrew Ng, but the questions, the portals and the programming questions are thought-provoking and intuitive. This actually gave a shape in my understanding in Deep learning which is an undefinable help for upcoming placement season and my M.Tech Thesis. Thank you, Coursera and Thank you, Andrew Ng Sir.

One more time Andrew Ng makes understanding complex things attainable. It's amazing understating and building Deep Neural Networks by applying all theoretical concepts and math. Indeed, this course leads you to reveal what's going behind the scenes in Deep Learning, instead of just calling "magic" APIs from modern frameworks. I strongly recommend this course. Background in Linear Algebra, Calculus and Python helps a lot.

The lecture videos are top-notch, very well explained and course materials are readily available. The discussion forums are very active as well and have been tremendously helpful for me. Would recommend this course to my peers in university as well. This course has been exceptionally helpful for me to gain some more insight on Deep Learning and NN as a whole since my minor specialization lies in AI & Intelligent Systems.

Andrew Ng is as always a very good teacher. The course is quite easy and I recommend it mostly for beginners. If you are already familiar with the topic you can certainly pick up some interesting tips but it is better to watch videos at 1.5 or skip through the content . The programming exercices are interesting, but too simple (there are only obvious one-liners to fillin) and can take a little less time than displayed.

Great introduction into how deep learning fundamentally works and how to implement basic neural networks yourself in Python.

The course even goes so far as to require essentially no prior knowledge of calculus which for my taste is already close to oversimplification, but allows virtually anyone to be able to follow the course contents. Because of that I can understand and support this decision by the course instructors.

This course wil not only help you understand the concepts and codes behind neural networks but also helps to visualize them in your head which is truly amazing . I always wondered how gradient descent works and now I have implemented that which is super cool. And in school or colleges no one told us what is the use of derivatives , well in this course I truly understood the essence of Chain Rule which just made my day.

Very good course. Considering my math skills are like 15-20 years rusted and I didn't know any Python before the course I felt very well mentored and lead through the programming step by step. The amazing thing, as Adrian also says, is that it's not in the programming, the magic is in the data.

Although I still don't fully understand where that Schroedingers cat/no-cat is hiding in the pixel data IT WORKS amazingly well.

Before doing this course neural networks seemed an impossible concept to learn .but along the way with Professor Andrew great way of explaining the intuition behind complex aspects and making them so simple and the high quality assignments.I can now say with confidence i know about neural networks and even can implement them from scratch using a python or any other programming language.Thank you Coursera for this course

As a person that's already studied deep learning at the doctoral level, this course still taught me so many things I didn't know. This is an excellent course for intuition and understandin the practical aspects. And Andrew is a very charming instructor that introduces the material in a very gentle and motivated fashion. I apprecaited the side commentary on we don't do certain things and origins of certain conventions.

This course was superb. It actually delves into all the nitty gritty mathematics behind neural networks and taught me how to build a NN from scratch. This unlike so many other online courses that just import a ready model and instructs you to just call .fit and . predict methods. Andrew Ng is a great teacher and explains all the difficult concepts in a clear easy to understand way. I wish I had taken this course sooner

The course was great and I learnt a lot from Andrew NG, one of the course instructors.

It's really a great course and I would recommend this to everyone aspiring to gain an insight into deep learning and neural networks.

The only thing worth improvement is the robustness of the assignments and to include and extra lecture at the end of the course to guide the students what sort of things to do next.

Thanks,

Suraj Upadhyay.

I am a Chinese student. It was very hard for me to learn a completely English course at first, but after two weeks, I have fully adapted to it. This course is not so difficult. Professor Andrew talks in great detail. If you are interested in AI, you must keep on learning, and you will certainly get something. By the way, be sure to make good use of the forum when you are learning, you will get a lot of useful answers!

The course video lectures are perfect to grasp the basic understanding of Deep learning and what it really is about. However, the assignments should be made more flexible and not bounded to specific format as provided. The students should think themselves and create the codes rather than giving a pre-defined format. Nevertheless, the course is indeed excellent for students to have better understanding of deep learning.

This is an outstanding course which covers the concepts and mathematics of Neural Networks very well. Professor Andrew Ng is an outstanding teacher and the way he explains and clarifies all the concepts, techniques and the mathematics involved is simply outstanding. Moreover the programming assignments are excellent and really provide detailed and precise instructions on developing each of the main parts of the system.

This lesson is excellent! The first time I see this lesson and Machine Learning I decide to learn it no matter how difficult it is. I do not major in computer science but the knowledge related to Machine Learning and Deep Learning is not so hard to understand. I feel ashamed that I have not found this lesson earlier. I will try my best to learn all the five lessons. After all, the subject I major in is called Big Data!

I really enjoyed and benefited from the simplified presentation of neural networks in this course. The method of calculating partial derivatives via the computation graph was particularly instructive and brought me one step closer to really getting my head around the mathematics of backpropagation. Future endeavors with more mathematically detailed presentations will likely bear more fruit for me thanks to this course.

Andrew manages to teach the processes of deep learning in a way I've never seen before. Instead of just being able to do operations featured in the course, I now have an intuition about taking things a lot further and into different fields.

More importantly, I'm really excited — it's been so good, I completed the first four weeks in just over one.

Thank you very very much to all who have facilitated this specialisation.

One of the best course for deep learning if you know a little bit about it and are familiar with python, don't worry about the mathematics, although it has been covered in the course, it is optional and in such a way that it's totally fine if you want to skip it, but according to me one should at the least try to understand it, luckily I am from a mathematical background and therefore it was easy for me to understand.

I like the pace of the course: not too fast, not too slow. Professor Andrew Ng repeats the most important topics each week so it helps to remember things better. The programming assignments are relatively easy--you can reuse the functions you created in the previous week. But even with these relatively easy assignments, you create your own real neural network from scratch! Looking forward to starting the next course!

Incredibly well thought out introductory course to artificial neural networks. Particularly appreciated the intent to have the student build every function relevant for the construction of the neural network architecture on his/her own. The final assignment is extremely instructive and I am sure I will review it multiple times in the future, as it is a great way to dissect the principal steps in forward and backprop.