This is an excellent course that explains the basics of neural networks. Other sources present a vague idea behind the algorithms and often fail at explaining the essence, but this course goes straight to the foundation of neural networks with a combination of linear algebra and calculus - so eventually nn become a simple set of equations which are easy to comprehend. By far the best course I have ever taken.

Presents the crux of deep learning simply and understandably. I wish Prof. Ng would go a bit more in depth into the derivation of the backward pass equations, but the course was otherwise good. Programming assignments are a mechanical at times because they spoon-feed you the answers. I found it amusing that they assume calculus is beyond the target audience but regularly assume some basics of linear algebra.

This course is great for beginners in deep learning field. It helps in understanding the main concepts such as: how to design model, the main parameters and the hyper-parameters, how to test the model and the difference between the N layer neural network and the deep neural network. The instructor is perfect and the whole course is amazing. I will recommend it to anyone wants to begin in deep learning field.

First course of the Deep Learning specialization. I did it after the Machine Learning course. It is a well structured and progressive course that really explain what is a deep neural network and we build the algorithm in Python step by step. The programming assignments are well prepared and finally give you a good implementation of a neural network as well as a good understanding of how the whole thing work.

it was super super awesome...

i thought deep learning is out of my league , but he braked it to me like a piece of shit..

I am currently studying engineering so i had a sense of algebra and calculus...

even though he said its had i was able to understand it, as a result now i have a clear foundational idea of how AI works..its so thrilling and fascinating!!!!! Thank you so much for the course .. lots of love

It's simply an amazing course. No doubt it taught Neural Network but for me most importantly it was a CONFIDENCE BOOSTER, "yes i am building something. It's AWESOME".

In the beginning i was getting annoyed because i was unable to find many solutions but the discussion form have all the required answers only thing needed is patience and calmness to search. IF YOU HAVE ANY QUERY DISCUSSION FORM HAS ITS ANSWER

This is my first course in AI. I am talking about AI a lot with people and had a conceptual feeling about it. After this short course I realized that my feeling about it was not the reality. I did not understand deeply how much the data is important, and that the network is generic. I thought that you need to understand what are the feature that you are looking for before you give it to the computer.

Thanks!

This course allows for a very quick review of the basics of neural nets. The coding examples/programming assignments have a lot of training wheels on them, so it is easy to focus on the concepts rather than the mechanics of data loading, cleaning, etc. Some may not like this aspect of the course, but you can always take a look at the utils file to get the details. Thanks for a great course, Andrew and Team!

I enjoyed attending this course very much and have the feeling that I have learned a lot about NN! The course is very well structured, the assignments are extremely helpful to learn how Neural Networks with different architectures work and also how to implement them in Python with enough support to not get stuck during this sometimes complicated tasks to build an NN. I am looking forward to the next course!

This course simplifies the deep learning concepts yet retaining the core mathematical details to leave you satisfied and yet enough curious to study further on your own. It also empowered me through the carefully detailed programming exercises that I can implement the concepts on my own. Overview of python and helpful tricks were really appreciated as well. I'm motivated to take Stanford's CS230 course now.

Excellent course, especially the last programming homework. I really enjoyed the theoretical and programming part of this course. Although I could prove each one of the theoretical of this course there is one big question I have and this what is the rate of the gradient descent approach. It is obvious that the x-axis variables are decreasing at the rate of the derivative, however, this approach works well.

I have just one request Just take 3 layer NN with 3 or 4 hidden units in each layer and please visualize how the image detection with parameters takes place. What are exactly parameters doing also what exactly vectorization and code processes during matrix operations . It may give more idea about how exactly pixels are detected and how exactly parameters do for people from mechanical or civil background.

it's really help full course those who is beginner for those its a gold and for those who intermediate stage for those its a special revision in which you will get to know new things which you might miss during your learning

professor is awesome his explanation style is awesome and he deliver contains very easy way.

Thank you sir we will definitely tell over children that you are the real avengers on earth.

A great introductory course to Deep Learning, and as always, Prof. Andrew Ng makes it sound so easy!

A small suggestion from my side would be to maybe increase the difficulty of the programming exercises. Small things like removing the hints, would make students work much harder and learn from experience rather than just copying what's given in the comments.

Just a small suggestion, otherwise a great course!

Course is great:

What I liked:

-Lecturer. He is the best. It could be seen that he loves what he is talking about

-Very interesting assignment results.

What I disliked:

-Assignments are very easy and almost done for you

-Some math stuff repeated again and again. This course could be 2-week long (3 at most). Yes, this could be hard for people who dont know calculus and LA, but they should catch up on their oun.

As always Andrew Ng's way of teaching is as simple that anyone from not much technical or mathematic background can do this course easily. In this course I learned about the basic fundamental of deep learning and how to implement it with multiple number of layers and by using different kind of activation function. This course also teaches back-propagation In a very unique and easy to understand way.Thanks

A great course! Andre Ng is one of the best professors I have ever attended a course from (either live or online). He has a really magical way of breaking down complex ideas and notions, and explaining them in such clear, concise and crisp way that even people totally unfamiliar to the field can easily attend and follow. If only all the professors were as good as Andrew, students would never skip a class!

I've reviewed machine learning and neural networks in the past, but I felt this version to be the best yet. He introduces all the topics precisely and succinctly, but you will need to know and do calculus outside of the lectures if you want to figure out why some of the derivations come out to be what they are. Although, knowing and doing the calculus is not necessary to understand and pass this course.

What compliments can I add that many others have not already conveyed before? This course is SUPERB. With basic understanding of Linear Algebra (matrix operations like dot products and transpose) and Calculus (first order derivatives etc.) and good programming skills in Python (basic numpy etc.), one can finish this course, if they persevere. Andrew Ng teaches the course with gentleness and thoroughness.

After one year experienec with deep learning. I signed up for the fourth and the fifth courses. However, I thought I would just check the first one to refresh my knowledge, and I was schocked that I learned a lot of new stuff. Informative, simple explanations and insightful. Also the assignments are perfectly structured. WOULD RECOMMENED THIS COURSE ALWAYS. thanks Andrew and your team. excellent efforts.

Best online course I've ever taken. Andrew is such a great teacher and always tries to open the black box of Deep learning and break down sophisticated stuff into understandable pieces. Another thing I like this course is the selected exercises assigned for each week, with which not only did I understand the knowledge I've learnt better, but I was applying them to solve some real industry-level problems.

Professor Andrew's explanation about the intuition behind the mathematics of neural networks is one of the best I've ever come across. Its pretty easy to understand even if you're a beginner in the field. Apart from the videos, the programming assignments were fun to do, especially when you watch your loss decrease after each iteration :) . Overall, a very satisfactory experience, 10/10, would recommend.

Hello Andrew, I am a big fan of you. Learning from your every course. Very unfortunate that I can do that remotely only.

One thing that I want to mention - Can we have lecture notes on coursera, just like the way used to in CS229 that we can read before coming to next lecture. I found that that was very useful in understanding when things get harder.

Thanks hope we can improve coursera in that matter.

Before going through this course, I completed a few of fast.ai course vedios, I even read 1st chapter of Michael Nielsen. Went through Derivates, Chain rule , matric multiplication. Only in this course I could get a clear picture of NN and understand how each of math is applied. Andrew was amazing in his explaination. I really liked those parts where he explained the various ways a equation is notated.

This is a great introduction to deep learning. If you have taken Andrew's Machine Learning course before, this will serve as a nice review of the main ideas behind logistic regression and gradient descent algorithm and also introduce new notation and formulation for deep neural networks. You will also get to experiment with creating your neural networks in Python using different libraries such as numpy.