Chevron Left
Back to Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization

Learner Reviews & Feedback for Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization by DeepLearning.AI

4.9
stars
62,825 ratings

About the Course

In the second course of the Deep Learning Specialization, you will open the deep learning black box to understand the processes that drive performance and generate good results systematically. By the end, you will learn the best practices to train and develop test sets and analyze bias/variance for building deep learning applications; be able to use standard neural network techniques such as initialization, L2 and dropout regularization, hyperparameter tuning, batch normalization, and gradient checking; implement and apply a variety of optimization algorithms, such as mini-batch gradient descent, Momentum, RMSprop and Adam, and check for their convergence; and implement a neural network in TensorFlow. The Deep Learning Specialization is our foundational program that will help you understand the capabilities, challenges, and consequences of deep learning and prepare you to participate in the development of leading-edge AI technology. It provides a pathway for you to gain the knowledge and skills to apply machine learning to your work, level up your technical career, and take the definitive step in the world of AI....

Top reviews

AS

Apr 18, 2020

Very good course to give you deep insight about how to enhance your algorithm and neural network and improve its accuracy. Also teaches you Tensorflow. Highly recommend especially after the 1st course

AM

Oct 8, 2019

I really enjoyed this course. Many details are given here that are crucial to gain experience and tips on things that looks easy at first sight but are important for a faster ML project implementation

Filter by:

6176 - 6200 of 7,216 Reviews for Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization

By Thitipon S

•

Dec 11, 2018

Parameters tuning is ok to follow, it would be easier if you have numerical methods basics. But Tensorflow is not easy to deal with. Maybe it need a separated course. I will get through to programming assignment again to understand it clearly with tensorflow manual pages.

By Jiachang L

•

Jun 19, 2018

The second class on machine learning is still very informative. However, it's very hands-on and teaches me mainly how to tune learning algorithms to run faster. Hence, it's not very intellectually stimulating. Nonetheless, this is still a very educational course overall!.

By Iliyan N

•

Jul 12, 2020

The course is great. Andrew is one of the best tutors one could get.

The only reason I rate it with 4 stars is that the TF assignment is not updated to TF2. TF 2.0 with Keras really is a state-of-the-art framework and imho there is not much value in learning TF1 anymore.

By Robbin R

•

Feb 10, 2018

Great sequel to Neural Networks and Deep Learning. Relatively short course and the most relevant topics in Deep Learning are reviewed. You also practice with TensorFlow, a well-establish Neural Network programming framework that is widely used in academia and industry.

By Richard J B

•

Oct 28, 2017

This course had more of a getting-into-the-weeds feel to it, without as much of the broader conceptualizations that the first course had. I also submitted several queries in the forums without getting feedback. Still good, but I hope the following courses are better.

By Chen Y

•

Sep 5, 2017

Overall, this course includes many useful techniques of how to further improve the basic DNN. Just one minor point imo that the tutorial on TensorFlow may need to go deeper for those techniques mentioned previously in this course, for instance implementing batch norm.

By Asad A

•

Aug 17, 2019

Great videos but wish there were more per-lesson exercises that were there in Course#1 for this track. Also, the transition to TensorFlow was quite abrupt as the key concepts that TF uses are completely new and don't easily borrow from the much cleaner Numpy concepts

By Laurence G

•

Aug 11, 2019

Decent intro to tuning neural networks. I felt the parts on normalization and regularization could have gone into more detail, but perhaps the math was deemed too complicated. Labs are ok, but still a bit buggy despite errors being reported in the forums a while ago.

By CJ

•

Apr 7, 2020

This is another great introductory level course. Andrew covers a lot of very practical concepts. This course also builds well on the previous course in the specialization. The only reason I gave 4 stars rather than 5 was that the programing exercise still uses TF 1.

By Bharath C

•

Jul 2, 2019

A good theoretical explanation and good working assignments that impart basic understanding of different optimization methods, hypertuning methods and tensorflow framework. But, some mistakes in the tensorflow assignment in the script itself, needs to be rectified.

By Joe G

•

Nov 5, 2017

Andrew Ng presents a very organized course; I would have enjoyed actually iterating on hyper parameters to find the optimal set. Also, there are probably other optimization approaches that would enable simultaneous searching for an optimal collection of parameters.

By Sawyer S

•

Oct 1, 2018

Overall, very clear teaching from the Great Mr NG, only concern is that the markdowns in the assignment three coding assignments have some mis-aligned expected output from what is actually expected, so there is some confusion. Except that, all is great. Thank you.

By Rohini J

•

Apr 15, 2018

It was very helpful to learnt batch normalization, regularization and tensorflow. It definitely needs a lot of self study to learn about these topics for people who are not familiar. Some mathematical resources like links to pdfs and videos would be extra helpful.

By Jean-Michel C

•

Aug 5, 2018

I believe we should extend this course for another week to properly cover TensorFlow. We end up copy / pasting code in the assignment without fully understanding the entire code. Otherwise, the quality of the course is always good, thanks to Andrew :-)

Thank you!

By Francisco J R A

•

Sep 20, 2017

Super interesting course where you'll have to come back a few more times because of the density of the theory. It's overwhelming the amount of hyperparameters you need to tune, but it also makes it challenging and less boring to set up neural networks and models

By Gary S

•

Sep 13, 2017

There are a few errors in the programming assignments, which caused some confusion. Finding these errors was a useful exercise, but it would obviously be better to have a debugging problem or two rather than errors in the problem hints or expected results. :)

By Dan B

•

Jul 4, 2018

The lectures are great - but the Jupyter notebook assignments are hell, as they they crash frequently and most of the time spent on the assignments is invested in dealing with the notebook instead of the exercise. (The content of the exercises is great though)

By Ganesh M S

•

Mar 31, 2018

The quality of the information is awesome. There are some minor bugs in the assignment section. Even though you have submitted the right answer it shows that you have secured 0 marks in that section. Apart from evaluation bug this course it super knowledgable.

By Kartik c

•

Sep 5, 2017

There were a few mistakes in the output of the comments of the notebooks,Also sometimes my output did not match the expected output,still the assignment got graded correctly.Eg-The tensorflow notebook.I think it was because of the seed of the random processes.

By Vahid N

•

Aug 4, 2018

Well-organized course. I gave it a four instead of a five just because the Tensorflow HW is not as good as other HWs. There should be more comments and more examples. Maybe there should be two HWs on Tensorflow to give me the confidence that I have leaned it.

By chandrashekar r

•

Sep 12, 2017

I would rate this 4 for the following reasons:

1) Learnt all the optimizations.

2) Hyper Parameterizations

I would not rate this 5 for the following reasons:

1) Some more time could have been spent on tensorflow.

2) The assignments were just simple substitutions.

By Elpidio E G V

•

Apr 23, 2019

Great explanations on behind the scenes operations of optimization algorithms and general theory. Coming from a more practical background, it helped me grasp the concepts much better. I only wish the programming exercises were a little bit more challenging!

By Amir M K

•

Aug 27, 2022

Generaly it was good as expected! But the problem with this course was the programming assignment at week 3, where it did not include programming training for most it's content which where Hyperparameter tuning and batch norm and was all about TensorFlow!

By Ansh M

•

Jun 26, 2020

It was a good course, with giving a great detail on tuning the Hyperparameters. I personally didn't myself found it useful as of now, but the course was good, and can be recommended to other people to fine-tune their networks. Jumping on the third course!

By Brook R

•

Jan 12, 2020

Programming assignment was more difficult but the Course itself really built on the first course well. I struggled much less with the material and enjoyed it more. I also appreciate it being shorter despite having to restart because I had gone on vacation