Started this course 3 months back, but from past two weeks I sat for around 4 hours per day, to complete this course. The programming assignments may not seem difficult intitally, because Andrew provide the vectorised equations but what really boils down and deepens my understanding was how am I going to use it in my application. How I will build my own image classifier ? When I try to answer such questions then yes it was very very helpful to me. I am still in learning phase, a beginner, so yes course was difficult but it was manageable.

It was an amazing course and I truly liked Mr. Ng's teaching style towards this complicated topic. In my honest opinion, I hadn't seen so far someone who explained DL foundations in such natural way and running the “deep dive” at the same time. For those who want to start the specialization, this is the landing course and It is highly focused on the back-propagation (BP) algorithm. I recommend to sharpen your differential calculus skills to truly understand what the algorithm is doing, specially for the general case and programming tasks.

This course is amazing !

I'm so happy that I've completed this beautifully crafted course . The instructor is really good.The explanations and presentations are so clear and easy to grasp.

Before taking up this course,I had a feeling that neural networs are very hard to conceive and implement...but this Course made me realise that anyone with basic knowledge in coding(python) and linear alzebra can easily learn to model Deep Neural Networks.

I thank the instructor Mr.Andrew and Coursera for offering this amazing course. Thank you so much !

El curso es muy bueno, particularmente estoy muy agradecido con COURSERA, por darme la oportunidad de hacer los cinco cursos de la Especialización en Deep Learning con ayuda economica y permitirme tener acceso a este tipo de capacitacion y certificacion. Muchas Gracias…!

The course is very good, particularly I am very grateful to COURSERA, for giving me the opportunity to do the five courses of the Deep Learning Specialization with financial aid and allowing me to have access to this type of training and certification. Thank you very much!

Course if fantastic starters, taking a mathematical approach to the design of NN. Assignments and quizzes are good as well.

However, The format of downloadable course materials need to be improved. It would be nice to see all the documents in one file for a certain Week, instead of downloading files separately. Basically the download format of ML course was much consistent and good for quick referencing.

But nonetheless, 5 stars because above is just my personal preference which has nothing compared to quality and content of the course.

I wanna thank you for your beautiful and nice website and your great instructors, everything was good but in my opinion if some optional short instructive videos or reading sections about 'dictionary ' and 'tuple' were between videos could be helpful, beside that having some reading parts contain abbreviation of videos that is written by instructors could be useful for student to review in a short time and organized them in their mind would be helpful because details are always forgotten and they're need to be reviewed several time.

Excelente curso introductorio, la curva de aprendizaje es un poco elevada al principio del mismo, pero si tenes una constancia y muchas ganas de aprender es un curso excelente. obviamente es introductorio y segun tengo entendido, todo lo que aprendes a hacer aca, alguien ya lo hizo, pero no esta nada mal aprender las bases matematicas y estadisticas del Machine learning.

TLDR: Buen cruso con buenas bases de matematica y estadistica. Necesitas conocimientos intermediosd e python o algun otro lenguaje de forma seria para engancharlo bien

This is truly the best course for those who want to start learning Deep Learning. Our Instructor Andrew Ng , he is amazing!!! . The way he teaches all the concepts are really good , the programming exercises were really helpful.This is a well structured course right from logistic regression to implementing Deep Neural Network.

Overall I really enjoyed learning this course and will continue learning this specialization and apply my knowledge to real world problems. A big thank you to my Instructor Andrew Ng and Thank you Coursera team .

I've been interested in learning NN and ML for a long time and Coursera finally provided this opportunity for me to do it in a timely manner. The time was very limited for such a wide topic, however, I believe they deserve a 5-star for how they managed to benefit such a limited time in a very efficient way. Andrew Ng is one of the best teachers I've had. He's both very knowledgeable, explains the concepts in a simple language, and he's very humble at the same time! Looking forward to getting more courses with him and with Coursera ...

Excellent course that teaches you first principles of DNNs. Very systematic approach by Andrew to start with simple concept of a shallow neural net and building upon it to introduce the concept of deep neural networks. Even though with frameworks like Tensorflow and Keras, it is easy to "engineer" a neural network, without building it from scratch like taught in this course, it is highly recommended that people take this course to develop a better understanding of how the deep neural networks work and why they behave the way they do.

Pros:

Pragmatic presentation of fundamental mechanics of feed forward networks. In particular I appreciated the clean tutorial of the ndarray vectorized implementations.

Cons:

The one feedback I would give is that the coding exercises had a lot of hand-holding. For a specific suggestion: some of the "asserts" used for checking correctness give away the answer. I suggest refactoring the checks to be private methods invoked in the notebook but implemented server-side. That way they can be inserted in the code without leaking the solution.

Amazing stuff. I've been looking for a good introduction to Neural Networks, looked through a lot of tutorials and blog posts (of which there are multitudes these days, since Deep Learning is all the rage now) which only confused me more, and finally decided to take on a full-blown course. Turns out, once somebody like Andrew Ng explains this stuff, it's no longer mysterious and convoluted.

Note, that it's better if you're at least familiar with matrices and vectors from calculus before taking this course since NN are all about it.

Excellent course, good balance between theory and practice. The teacher thoroughly explains all the elements of deep learning before you're given the programming assignments. He gives you both the theory and the brief overview of how it all works. The programming assignments are designed so that you only focus on the "neural networks and deep learning elements", you won't have to worry about programming environments or what libraries to use, which saves a lot of time and gets you going on the most important aspects of the course.

Wonderful course.At the beginning it even seems to be too simplified (course team explains everything and structures the code for you). But this is just an illusion. Closer to the last week you start understanding that multiple reherasing of the basic neural network concepts is key for conscious understanding. And that structured code is wonderful (in Russia it's not practiced =(( ). Separate thanks for backpropagation explanation with computation graph. That was very helpful.I'll definitely recommend this course to other people.

This was a grate learning experience,I have not seen a single tutorial that has covered building Neural Network from scratch like this one.Perfect combination of code and the underlying concepts have been explained in a very intuitive manner.The additional part "Heroes of Deep Learning " was very much inspiring. The discussion forum was great I've learnt a few additional things from there. Thank you sir for providing such a quality course,I'm very much satisfied with the quality of content and as well as the method of teaching.

As someone with a strong background in mathematics and a good programming skills, I found the course level rather "basic" and I could quickly absorb all the lecture's material. I found the materials extremely interesting and well organized. The assignments, though rather straightforward (implementing what has been explained, nothing more) were difficult enough to made me feel I was "building" something. And then, the possibility to experiment and play with the code was also great. Overall, a very good corse, thank you professor!

Sometimes it's difficult to connect to the server when doing the programming exercises. The course is well structured. However, the programming exercises can sometimes be confusing because there are quite a few "helper" functions in the deep learning algorithm. Students may need to consistently refer back to the help functions defined earlier to implement the final learning algorithm. Overall, it'a great introductory course. Andrew has given very clear explanations and useful pointers when implementing deep learning in practice.

It was a nice experience with a such a experienced and well knowledge supervisor. Who just started right from the beginning and then distributed the course in easy chunks so that all the content remain understandable to every type of learner. I thanks specially to Higher Education Commission of Pakistan who has offered such a splendid course to increase our domain knowledge and also thankful to supervisor and coursera team who have done such a excellent offer to spread the knowledge by using most modern techniques of learning.

This course is my first in Deep Learning and has been very interesting for me. The inclusion of the notebooks and grading are a very useful touch. Andrew does a good job trying to abstract away the complexity of Deep Learning, but it still does require some understanding of programming (python), calculus (mostly derivatives), and matrices/linear algebra. For someone new like myself, I find that I often need to pause the video, take notes, and also just rewatch lectures multiple times before I start to understand the material.

NN&DL is much shorter and much easier than Dr. Ng's original ML course, and the material overlaps a lot. I finished the course in 2 days (not 4 weeks). As always, Dr. Ng's explanations are clear, and the material is beautifully organized. I felt the answers to the assignments were a little too easy. Not really a lot of thinking required. The interviews with the "heroes" of Deep Learning were fantastic. Their ideas about the future were very inspirational. I am looking forward to the next course in the specialization.

It was a wonderful course to get started with Artificial Intelligence and Machine learning.Those concepts of forward ,backward propogation, relu and sigmoid function was really new and helpful to get insight of what happens behind the scenes of machine learning algorithms many concepts were new and typical but Sir Andrew did a great effort and explained them in a way that everyone can understand it. I highly recommend as a student to take this course and challenge your skills with what you can do to contribute in AI world.

This course is a very good kick-start for learners in deep learning. Prof. Andrew Ng explanation covered most of the details required for building neural nets and the programming assignments gave a clear idea on working of the neural nets. I got stuck at some point in programming assignments, later I completed it successfully before the course ends. I experienced the same excitement from starting till the end of the course. Thanking coursera for also providing financial aid. Looking forward to complete this specialisation.

This was such a helpful lecture. It is very well organized, and great for all learners with various backgrounds. I was very surprised with the diversity of people who take this course. The discussion forum on this course is absolutely fantastic. You can find all the possible problems/questions you might run into, and they are all answered by numerous mentors on this course. I highly recommend this course to anyone that are looking forward to start deep learning/ai. I am actually anxious to start the next course materials.

I loved that this course married both a 'top-down' and 'bottom-up' approach. I started my deep learning journey with FastAI (not to slight Jeremy, he is a phenomenal teacher and I understand the logic behind his teaching style), however was craving some 'lower-level' concepts out of part 1 WITH math notation. I thought this course did a great job of finding a medium between these ideas: starting with something lower level + math notation, but also providing practical notebooks and algos with working model implementations.

The main take away for me from this course is to learn how to systematically denote various quantities involved in deep-learning such that they can be recognized later without any confusion (e.g. dW is gradient of cost with respect to W and so on..) and to learn how to structure a code to implement any deep neural network. Also, from data analytics perspective, I learnt about the limited representational capabilities of simple models like logistic regression and why deep networks tend to work better than shallower models.