Recommender Systems and Deep Learning in Python (Udemy.com)

Good course to get an idea of some of the most important algorithms / approaches within in this field. The introduction could be a little bit more general and more overview of the field could be given, such that the approaches are aligned within the overview. Regarding applying the approach to real world the course lacks information, also the usage of good evaluation metrics is poor. A policy on slide and code usage could be added and I find the approach on how to get the slides too rigid as well. Generally the lazy programmers courses are based on the foundations, I think he repeats this too often in his courses being afraid of bad reviews. More value on intuition and the foundations could be added here (e.g. adding addendum of important lecutres of other coures), yet most of the time he gives a good understanding of the most important concepts.

Especially liked the deep learning portion. Wow! This is simply one of the best courses out there for recommender systems. It covers every aspect from collaborative filtering to Google PageRank to Bayesian methods to matrix factorization and big data. You'd learn how to build the model and the workflow. You'd get to work on exercises to reinforce learning. This course certainly meets my expectation, and it's totally worth it.

I only completed the first 3 sections, which already gave me what I needed to progress on my personal project. I will certainly be back for the DL approaches. I personally like this instructor a lot and would definitely recommend it, but only to those who are willing to put in some work. Completing these courses is slower and more demanding than most other udemy courses on data science, but the content is of high quality and teaches in a solid and long-lasting way.

The structure of all topics is great. the codes are explained clearly. The thing I think it can be improved is about math equation explanation. I have seen from other instructors that made me can follow thought. (However, I also like that you have the section of math equation explanation, just it can be better than this)

This is a great great (not typo, I said great twice) course. It is far better than any other courses that I have taken about ML, DL, DS, etc if you know enough math. It is definitely the best choice if you want to know not only how to practice these methods, but also the math behind them.

The sequence of lectures built nicely on one another. It goes from simple recommender systems to collaborative filtering and matrix factorization, then to advanced deep learning and big data. Maybe you can't make it through the whole course in one go, I think that's a good property. You can keep coming back to the course and finding more and more value.

The presenter is clearly very skilled in this subject matter. He has put a lot of thought into how to present the course. Nevertheless, recommender systems are complex by nature. YET, simple to conceptualize. For the best effect I suggest you listen twice to everything, be curious. He moves fast through many topic areas, so don't be afraid to see how he has interpreted things and conceptualized a solution. I am still getting used to Python, it data structures and conversions. Listening twice allows me to see and understand the necessary transitions and manipulations.

Lazy Programmer is a rockstar. If you take seriously this course is equivalent to doing 3 weeks in college. The course is not only implementing the coding part but also deriving the math and bayesian statistics behind the models, which is absolutely amazing. He doesn't have energy in his lectures and it can seem boring at first, but it's worth to check the entire content up to the advanced parts. 5 stars.

I bought another Udemy recommender system course but that course made me crazy because that course just throw me names of theories without deep explanation and the programming projects used Surprised package, which taught me nothing about the algorithms and made the things hard. While this course is very challenging, the instructor does a good job to explain the theories and teaches me how to build recommender system without packages. For this course, Lazy Programmer is much better.

The course is very good overall. Lazy Programmer demonstrates for another time his advanced knowledge in machine learning. Personally, regarding matrix factorization, I would expect him to work towards the improvement of the results (and not just leave them as such) where he would also inevitably provide some other hints about how to efficiently tune these algorithms and how these work under the hood. Also it could be also useful to use at some parts of his matrix factorisation any relevant library to recommender systems (e.g. Scikit Surprise) to make people more at-ease with them and perhaps also provide his view on the implementation of the relevant algorithms at these.