Unsupervised Deep Learning in Python (Udemy.com)

Behind the complex mathematics and statistics, this course gives a very intuitive and comprehensive explanation about all of the subjects and gives some approaches to solve several problems. The course is exhaustive in describing loads of algorithms. I wish there were more DIY codes and exercises. They are making it too easy by giving you everything. Cheers!

I have done a few courses in this series now and all have been great. This one has proven particular useful/interesting to me though. Concepts are clearly explained and there are plenty of opportunities to practice what you are learning. In particular the motivation for using unsupervised learning is now a lot clearer to me than it was before taking the course.

I absolutely loved the theory part of this course, in particular, how the instructor got into the details of PCA and RBM. I have done many ML courses, and I have never seen anybody else go into the mathematics in such detailed. Loved it. The coding part has some minor annoyances, end of lines not visible in code along and instructor not explaining himself while writing the code. This improves a lot with the extra lectures that have been added to the course at a later stage. However, this only give you an incentive to do the implementation by yourself and only use the lectures as a check. Not enough to deduct anything from my score. This is my second course from the Lazy Programmer, and it surely will not be my last.

Brilliantly taught course (as usual). I loved this course specifically because it was more inclined towards mathematical derivation of the results and then applying the results in form of code. I did search a few research papers initially before enrolling for this course and the content was too vast to start from. Now, I have got a direction as to where to go next and what all things I need to explore from here on. I took this course too seriously and have spent around a month on the course writing notes wherever possible (even for the codes). So overall I can say, everyone should at least do this course for better understanding of the concepts.

Good explanations, regarding autoencoders: stacked autoencoders and deep autoencoders. Also, RBM (Restricted Boltzmann Machines) are well explained. I give 4 and not a 5 because it was what I expected, but was not above my expectations. To be above my expectations topics like sparse autoencoders would need to be covered and perhaps other topics.

Easy enough to follow. The RBM part is just copied from the Recommender systems lectures. Lazy programmer indeed :) Through this course I understood t-sne better, and figured out how to use auto encoders. It is a useful course, and was well worth the 9.99 I paid for it.

Does a great job of explaining the intricacies behind unsupervised deep learning and why they are applicable in the real world environment. I enjoyed the exercise for coding from Tensorflow each time. It's great to understand how everything is working so well. I believe it's a must have for anyone who wants to know about deep learning.

This course provides an insightful look at the theoretical parts of deep learning. It gets to the core of some concepts like vanishing gradient problem and how autoencoders and boltzmann machine can be used there. The course gives a mathematical justification for all the algorithms which helps me with my understanding. It uses Python, and you can implement whatever you learned straight away.

This is a unique course which provides mathematical insight into deep learning and neural networks. The videos provide very clear explanation and intuition behind unsupervised deep learning methods. The most helpful part of the course is seeing all the code. The examples gave a good understanding of the implementations and how to build neural networks piece by piece.

This instructor is clearly knowledgeable and well versed in the topic. The one place where I though my time could have been spent better was during the lectures watching the instructor type code. What I think would be more effective and time-efficient is perhaps displaying code and adding annotations and talking through the motivation of each step. Great intro to this topic, by the way and I shall look for more courses from this instructor.