Chevron Left
返回到 Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization

學生對 提供的 Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization 的評價和反饋

56,632 個評分
6,495 條評論


This course will teach you the "magic" of getting deep learning to work well. Rather than the deep learning process being a black box, you will understand what drives performance, and be able to more systematically get good results. You will also learn TensorFlow. After 3 weeks, you will: - Understand industry best-practices for building deep learning applications. - Be able to effectively use the common neural network "tricks", including initialization, L2 and dropout regularization, Batch normalization, gradient checking, - Be able to implement and apply a variety of optimization algorithms, such as mini-batch gradient descent, Momentum, RMSprop and Adam, and check for their convergence. - Understand new best-practices for the deep learning era of how to set up train/dev/test sets and analyze bias/variance - Be able to implement a neural network in TensorFlow. This is the second course of the Deep Learning Specialization....



Thank you Andrew!! I know start to use Tensorflow, however, this tool is not well for a research goal. Maybe, pytorch could be considered in the future!! And let us know how to use pytorch in Windows.


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


5601 - Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization 的 5625 個評論(共 6,423 個)

創建者 Amir H


The explanation and examples are very informative throughout the course. The quizzes and the assignments are highly related to the topics covered in the videos which provide a solid understanding of the course.

創建者 Luca V


Some very interesting consideration, though I would have liked a section about reproducibility and randomisation (including for GPU trainining), though I understand that this is framework and language dependent

創建者 Karl M


Some of the programming assignments are a bit confusing, and the grader seems to suffer from bugs at the moment. Nevertheless I found especially the part on optimization algorithms very helpful and interesting.

創建者 Baris K


Maybe TF should be thought a little earlier with small exercises in the weeks 1 & 2. Also the final programming assignment should be improved. The seed initialisation at the Xavier initializer is ambiguous.

創建者 William R


The insights and intuitions Andrew communicates are good, but as he starts to point out towards the end of this course, in practice one uses a DL Framework and you don't code these things from the ground up.

創建者 Martijn v d G


The level of detail in this course really leads to a good understanding. A bit more programming exercises with TensorFlow (more than a single model) would be good to understand the intricacies a bit better.

創建者 Armaan


Extremely well designed course, the key reason for 4 stars is Andrew Ng's amazing leactures. The programming assignment though do quite a bit of handholding which can be reduced.

Amazing experience overall!

創建者 Haiwen Z


The course is great for beginners, and I'll recommend watch the vid with Deep Learning on MIT Press. The only cons for me is that subtitle is toooo big, I wish I can change the font size on the vid setting.

創建者 Gianluca M


Very short, but very interesting. Some more advanced topics are presented that students don't typically learn on coursera courses, such as improvements to gradient descent, batch normalization, and dropout.

創建者 Philip


Good course, not quite as intuitive as the first course in the specialisation 'Neural Networks and Deep Learning' but still very good. Its also great to have some exposure to Tensorflow through the course,

創建者 Arsen K


Great course. One star was taken off, as I would like to see more in-depth info on Batch Norm and a bit more discussion on how to compute gradients in case that is used. But generally that's a minor detail

創建者 Jatin


This was a great course....but at some places I felt that the details have been hided a little....only in few videos.........but overall it was a great of the courses...I have ever seen ..

創建者 Ashwin A R


This course helped in deepening knowledge about optimization techniques and how they could make ML/DL algorithms robust while training. This also provides a good introduction to the Tensor flow framework.

創建者 Charles H


The lectures are all really good, but the programming assignments feel like they hold your hand too much. It's very easy to sort of slide through them without having a good understanding of the material.

創建者 Aditya K


Everything till now was good, But I can't tell why my forward propagation method is rejected although it matches the expected output. So my marks were deducted for it without any reasonable explanation.

創建者 Vu N M


A bit boring with this course at the first sight, but later when you work with the real system, this course can be a bible for you. The valuable experiences from Andrew Ng are shared through this course

創建者 Gillian P


Though very good, his course might be a little less polished than the previous. One more week diving into frameworks would (maybe keras to see a more functional level of Framework) would be appreciated.

創建者 Manoj A


There was no exercise on hyper-parameter tuning so the course seemed incomplete. I think week 3 should be split into 2 weeks with first week focusing on hyper-parameter tuning and second on TensorFlow.

創建者 Øystein S


Ng is an excellent teacher, and it was fun to learn about programming frameworks. However, the programming exercises are very simple, and the videos about numerics go very slow, thus 4 stars and not 5.

創建者 Benjamín V A


Very good course, useful and smart. Some of the example are on tensorflow 1 but I think that they will update them soon to keras tf2 Thank you!

I will pass on what I have learned here to undergrads :)

創建者 Yan L


very useful course, especially the last tensorflow assignment. the only reason i gave 4 stars is due to the lack of practice on batchnorm, which i believe is one of the most usefule techniques lately.

創建者 Ashim


Assignment in week 2 could not tell the difference between 'a-=b' and 'a=a-b' and marked the former as incorrect even though they are the same and gave the same output. Other than that, a great course

創建者 Hans J


great and practical insight. carefully crafted assignments. still coding in python and the quirks coming with it are sometimes of equal difficulty if not worse than understanding the explained theory

創建者 Kevin C


Excellent content. The grader seriously needs to be updated thogh. For example, it needs to be Python2 and Tensorflow2 compatible and also needs to be robust in handling common syntaxes such as "-=".

創建者 mitch d


Would have liked to see the math and more complete explanations for all the things that Prof. Ng glosses over by saying "you don't really need to understand XYZ". Even if this material was optional.