For those who doesn't sure about if they meet the prerequisite:

Very basic of linear algebra.(concept of matrices)

The concept of calculus.(chain rule etc.)

Basic programming language to start.(for loop, iteration) and it would be very helpful if you've learnt python.

So, I think this course is very easy to take, especially English is not my mother tongue, I still can understand this course, thank you Andrew and all of the mentor!

After paying upwards of $1500 for an Intelligent Systems unit as part of a CS degree I still found even the basic concept of neural networks to be confusing and intimidating, despite a final grade of just over 90%; this course not only cost a fraction of the price, but explained the content matter and surrounding ideas in a fashion which made it much easier to digest and laid a strong foundation for further learning. Thank you!

Prof. Andrew Ng has done a wonderful job in this introductory AI course (Neural Networks and Deep Learning) that is easy to follow for someone like me who is new to AI from input data, initialization, model building, train, to prediction. You will learn the key concepts like forward/backward propagation, cost/activation function, gradient descent, parameters, etc. that work like a magic to recognize pictures has a cat or not!

It is an incredible course, probably the best one I have ever took in Coursera. I have learnt things I thought I would never do. Andrew is an incredible profesor. I highlight that he explains the math around neural networks and not only their implementation; that was my favourite part of the course. The programming assignments are very well prepaired and help you understand and internalize what you have seen in classes. 10/10!

Initially I was highly unsure whether or not take this course. But i feel completely different after my first course finished. The course is supremely organized and details the minutest details for deep learning from the very beginning. The hands on over the assignments has made things more clear. Hope i will finish the other course in the specialization soon and share the overall feedback. So far its the excellent experience.

Even some of the materials are too easy, it is a super great course for beginners without too much knowledge in Deep Learning and Linear Algebra. I highly recommended people to take this course and be patient about the repeating emphasis on vectorization and etc. that might make you feel boring and redundant. I personally gained a lot more knowledge and skills by successfully complete a multi-layer Neural Network step by step.

It is a lot of overlap with Stanford ML course. I like introduction to Ipython Notebooks, Numpy for the programming assignments in this course. I enjoined interviews with the people from the field of ML. Compared to Stanford course I think there is less reference material you get after you finish class. So I recommend to take Stanford ML course, if this is your first ML class. Besides doing assignments using Octave is fun too.

I initially knew a lot of theory regarding neural networks, but had no real experience in coding them. This course not helped me refresh my knowledge of neural networks but also gave me the confidence to code them from scratch. The implementation and usage of helper functions really helps improve the code for the neural network and helped me break down the problem to sub-problems which made the coding effort less intimidating

The explanations and content of this course are excellent. Though previous python knowledge is required and a bit of previous machine learning knowledge is recommended, this course provides the information in a great way for both begginners and intermediate learners on the topic. The Jupyter Notebook assignments benefit greatly by providing interactive assignments with clear explanations and instructions in an ordered manner.

The course gives a very thorough introduction to the concept of neural networks. Even though it's very easy to implement neural networks by using frameworks, the assignments in this course helped me to understand much much better how networks work 'under the hood.' As an engineering major, I felt there wasn't enough math. However, the course gave me enough intuition to understand more mathematically involved texts by myself.

Great structure of the course, from really simple NN to deep NN. It was very easy to follow the train of thoughts and learn the basics of deep NN. I also liked the interviews after the end of every week, I believe that help us to motivate ourselves. Finally, the programming assignments were well-organised and helped to fully understand what we learnt on lectures. I'm very satisfied by completing this course1 of specialization

An amazing course. The enthusiasm of the professor is clearly conveyed in his gestures and the hard work put into the production of the course videos. The weekly quizzes serve as an excellent way of both ensuring that you understand the theory and brushing up on it at the same time. The programming excesses serve as excellent tools to obtain a working understanding of the implementation of normal feed-forward neural network.

Excellent Course! I like that the lectures are not too mathematical, as some time has passed since I graduated. I liked the exercises, but I'm not sure why we implement our own network code. Wouldn't it be more practical to use TensorFlow etc... for the exercises? Also one open point: I still haven't understood why we use the log-loss function for logistic regression (and in all our examples) and not the squared-dirrerence...

I started leaning AI, Robotics, Neural Networks, Fuzzy Logic and Lisp between 1993 and 1996 in an attempt to achieve graduate degree. I was not able to get student financial aid and I ran out of money. I quit school and found jobs as contractor. I thought I should study this again and it would be easier this time. I hope to continue in this field. Thank you, Coursera, Stanford university, Andrew NG and his team.

Tareq

Before taking this course I was graduated having only basics of machine learning and NN.

I learned a lot from scratch in this course about Logistic regression, than NN introduction till detail and ends with implementing my own DNN, writing forward plus backward propagation in python using numpy.

It's really good for beginners who are really passionate about deep learning and have very basic concept of machine learning and AI.

Solid introduction to Deep Learning with excellent videos from Andrew Ng. Found the videos from Geoffrey HInton and Ian Goodfellow very useful for setting the context of the class and for helping me consider areas of specialization. I'm wondering if one or two additional application exercises (besides the cat ID) wouldn't have been a nice touch. In any event, I'm excited subsequent classes in this Coursera Specialization.

Great course for anyone who wants to learn about neural networks and deep learning in great detail by using mathematical operations and basic programming tools instead of pre-built libraries such as Keras or Pytorch. One downside of the course is that it may be difficult for people who are not good at math or programming in order to grasp the essence of the presented formula; however, it depends on your personal preferences

Before starting this class, I was trying to learn tensorflow, and it was going great, but I felt like I was missing the core reasons and the understanding behind why they had to model the networks a certain way, and what were activation functions and optimizers and so forth. After taking this lecture I can say that I have learned a vast amount more on Neural Networks than expected and I can't wait to start the next course.

An excellent introduction to the basics of neural networks with the optimal amount of math to build a solid foundation for further courses in this specialization. Don't be deceived by the lack or simplicity of "real" math in this course. Further courses will contain more complex math which may be difficult to understand without a background in linear algebra and calculus unless you have carefully followed this first course.

i am very grateful that the team of neural networks and deep learning have given me scholarship so that i could learn the basics and then could move ahead for the advanced version of deep learning .

This course provides the necessary base for the development of the understanding the difference between the neural network and logistic regression. And the assignments provides the necessary boost need for success in this course

After taking Andrew's first ML course several years ago, I felt I needed a refresher. This was really challenging but as I progressed through the labs my understanding solidified. My only advice is you should have at least an intermediate level of skill with Python and a bit of experience with Numpy before taking this course. With that said, you'll be receiving a practical, world-class AI education from the best, Andrew Ng.

Course videos are clear, to the point, and nicely segmented. Many comments in the video impart wisdom acquired from years of experience a deep learning pioneer. Programming Assignments that solidifies deep learning concepts are nearly perfect with clear instructions. One exception is the "L_model_forward" function in Week 4; the Grader requires caches be stored with exact indicies, which isn't mentioned in the instructions.

This course was quick and concise, and it is by far the simplest way to understand a very complicated topic. I've been studying AI and neural networks for over a year now, and this course still managed to make some things click for me. It's amazing how differently you learn when you're forced to implement these networks practically by hand, no tf or keras. Great job Professor Andrew Ng and the rest of deeplearning.ai staff!

I am very happy with this course. It gives you a better knowledge of what are under the hood in neural networks. I am looking forward to the next course in this specialization.

Estoy muy contento con el curso. Este curso te da un mejor entendimiento de como funcionan por debajo las redes neuronales. Estoy deseando empezar el siguiente curso para saber mas acerca de los hyperparametros y mejorar el rendimiento de los modelos.

I was able to complete this in a weekend. I went from knowing nothing about "deep neural networks" to completely grasping the idea and theory behind neural networks. It is clear that a lot of care was put into perfecting the notation. One remark: at some point we decide to call the aggregate of w.T as W. There's obviously good reason to do this, but I think that should be stated more explicitly. Wonderful course Andrew Ng!