Deep Learning Prerequisites: Linear Regression in Python (Udemy.com)

The Effective Learning Strategies section was important to manage expectations of the course. It's where the author summarizes how the topics are presented and the background knowledge necessary to understand them. I wish I had seen this first, along with the prerequisites, as it would have allowed me to focus my learning better. Author is clearly knowledgeable about his subjects. There are some places where I've learned more in 15 minutes from one of his course sections that took 15 hours in others I've taken. Overall, I'm pleased with the content, pace, detail, and reasoning presented here. Starting with the gradient descent section, though, the pace was more hurried and didn't supply enough explanation for me; maybe because it was taken from the "Unsupervised Machine Learning" in a different course. A helpful resource to me was the Wikipedia topic "Gradient". Also, without the assumed exposure to matrix calculus, the content became overwhelming instantly. Fortunately, the lessons were progressive, I refreshed my knowledge of math rules from other resources, and I was able to eventually absorb the math theories that the algorithms were built to implement. The coding examples that relied on existing libraries were pretty straightforward to me, since I am more programmer than mathematician and reinforced how theory was being applied. That said, I still had to repeat parts of the course multiple times, deriving the formulas by hand, before I could recognize why the solutions worked the way they did. It should take less time for someone with fresh knowledge of the mathematics involved. Begrudgingly, it did feel good to reacquire those skills and the concepts made more sense afterward. I'm being picky, but author's opinions of "obvious" and "dead giveaway" were sometimes neither to me. Hearing those words was like the discordant moment when someone writing on a chalkboard made the chalk screech. It was simply a distraction that interrupted learning.

This course presents linear regression and gradient descent which are fundamental tools in statistics and machine learning. The instructor describes concisely and precisely the math theory underlying them and how they are related to ML . He also applies this material to several examples using Python. At the end of this course, the learner should be able to apply linear regression, multivariate regression and gradient descent to various data sets assuming he can sort them. I recommend this course to everyone interested to learn these tools and to apply them to ML or other research fields.

Excellent course. The beginning was good, the lecturer explained the basics of linear regression well, with examples and at appropriate pace (at least for my background...) What really stood out to me was Section 4: Practical ML Issues. It was super interesting to see the connection of linear regression to statistics, the prior/posterior distributions, etc. It didn't *need* to be a part of this course - I assume many such courses just explain the basics and show how to code it up - but here it really helped me to gain a more thorough understanding of the topic. Thanks and now onto logistic regression.

Lazy has a very profound knowledge about the subject is easy to see. Didactic on other hand is not his priority.

I think you can do better.

This course is exactly what I was looking for. The instructor does an impressive job making students understand they need to work hard in order to learned. The examples are clear, and the explanations of the theory is very interesting.

Unlike other courses, the theory is well explained. However, a more mathematical explanation in maximum likelihood, map and regularization would be more helpful.

Lot of repetition and sometimes awkward sequence of topics.

Except the probabilistic part , everything does Good!

Best of the best

In the beginning lecture, the course seems fairly homegrown. Information is good, but there is a lot of background noise in the videos making them seem less than professional. Later lectures are better as far as less-to-no background noise, but you can tell which lectures were done originally and which have been re-recorded by the volume/background noise.
The information is good and what I was looking for - in-depth information on how linear regression works at the mathematical level. I had already taken a few machine learning classes before this one.
I would suggest starting with some of the lectures in the appendix before jumping in to the course material and wish I had known to do that. Some of the "Extra Help" sections would have been more useful to me up-front.

prediction of stock movement