Machine Learning Applied to Stock & Crypto Trading - Python (Udemy.com)

A good primer for getting a basic understanding of some machine learning concepts. I personally do not like to watch code being written and would rather have more examples with time spent fine tuning and improving models. Some of the libraries used here are deprecated, like gym, or do not work like DataReader with yahoo finance. There is a bug the code in the final workbook (action variable is never set) which is not picked up and the results are not as expected which is explained away incorrectly. This is picked up by another student in the Q&A but there is no response from the teacher or subsequent correction of the course material which leads me to conclude that the course is no longer actively monitored or updated.

Excellent course! This course is designed for traders, offering a very applied execution with the proper amount of theory. Whenever I felt lost, the instructor provided resource recommendations that helped me get back on track. I discovered this course through his Rust GPT course. I was able to use Polars and NumPy as alternatives to Pandas throughout the course. Working through the material enabled me to learn the application's and limitations of Polars when interacting with Matplotlib, PyTorch, and XGBoost. Essentially, whenever Polars wasn't suitable, I used NumPy arrays instead.

Best course on machine learning with trading I've taken on Udemy. If I had to nitpick then I would say I would have liked to see more explanation on best practices for adjusting the rewards in the reinforcement learning section. I plan to learn more about RL and circle back to the strategy to try to improve it though.

out of many courses I have gone through related to machine Learning and AI in Finance, this one stands out to be the best. I suggest people to have basic idea of trading, building technical indicators/strategies, ML and AI concepts so that you can understand the level of information he has provided , that can be truly implemented in real time for making profitable algo bots. Thanks for providing such wonderful course. Only thing I would request further (not just here but 99.99% of courses) to show us how to use ML models, AI models with real-time data, like, you can connect to any test-net and show us, how to run the models which we have developed and predicted to be used in real-time. This will truly help. Thank you.

Very easy to follow from start to finish and this course has opened my eyes to the different models I can use for Machine Learning. I have a project that I want to code and after completing this course I now feel confident that I can choose, code and test the correct model for my problem. Fair warning, the reinforcement learning section is heavy and needs full concentration and replaying parts of the lectures to get through. For me, that's OK as I am going XGBoost! My next step is to take an XGBoost specific course; this is just to build my confidence as I could, probably, build a model from this course alone. In summary: I am very happy that I purchased this course.

I feel humbled that someone would share this information with the world. Most successful people keep their IP to themselves, whereas Shaun is sharing it. For me personally, it's been the catalyst I needed to enter the ML Trader world. I've been trading for 6 years, and feel its time for ChatGPT to help automate my strategy. With their being so many options out there I became overwhelmed and afraid that I was allocating my time to the wrong models. This course has helped resolve my once challenges, and added answers to questions I didn't know I had yet...'as such' ;)

As a trainer, programmer and trader, I will state: This is the best course (probably) on Udemy on the subject. The instructor is a hands on guy who just gave a tip of what he really does and not what he read. He touched on major topics that are most critical to trading - regime detection - market asset clustering (segmentation) - Dimensionality reduction - Supervised Learning. But because the course is so hands on, to get the best of this course, - one has to have known the theory around the algorithms and what they really solve. Example, I have done many types of market regime detection, so I appreciate one when I see one. - go through the materials at least twice. Once to get what the instructor is trying to do. Then doing the same yourself. Then with other instrument. What I would have love to see also here, is model deployment. Nothing is perfect but model deployment would have pushed this course into the realms of perfection.

The instructor is extremely knowledgeable. One of the most knowledgeable on the topics. However, following along was difficult. I either had to open up his completed notebooks to see where he was going with his code... Or I had to watch a few videos all the way to the end before I understood what he was working towards... And then start from the beginning again, knowing what he is coding towards. That is the only reason I did not give a 5 rating. But the wealth of knowledge contained in this course is mind blowing.

Just Awesome detailed oriented. Would help frequent output review to understand typed code. Thanks to Shaun for putting it all together. Shaun is my first choice when I'm looking for something new to learn. He has courses on niche subjects and highly good at presentations. Typing and reading out loud is not what a course is all about. Although Shaun does explain certain code-lines, the majority of them seem to be skipped due to whatever factors only Shaun knows ;-) The essence of any course is unfulfilled unless each line of code is at the least briefly explained why, where, what, which etc are the reasons for it's existance and use-case.

This is kind of a knowledge that should never see any sunlight, you know. It is a secret of large corps and intelligent financial engineers. You feel like you must be quick to act before it becomes common knowledge after you learn some of the concepts taught in the course. o.O You will get angry at yourself when you don't fully understand what's going on but keep learning and you will achieve it! That being said the course could have been much concise because the instructor teaches you how to run his code most of the time instead of telling why things work that way. So he typing the code part is pretty much unnecessary. You are expected to go and learn bunch of side concepts on the go. But go don't really need to in order to achieve your end goal, which should be building your model with trial and error, if you're not a machine learning data scientist.