This is the repository where I keep all of the work that I have done while taking online courses. They are organized by the course provider.
A machine learning course devoted to understanding the mathematics and the machine learning algorithms themselves. Learned how to implement most of the most popular algorithms. Although I learned a lot of interesting information and I came away with a better understanding of area of machine learning as a whole, I was not able to turn around and really apply what was taught. We were handed complete and clean datasets without any notion of how to preprocess data. It is also not a good idea to try to implement the machine learning algorithms yourself outside of just trying to understand them yourself or you are doing your PhD research into machine learning. The libraries that have already been written have many many professional and academic eyes on them and are most likely written correclty and as efficiently as possible (up to this point). I don't think someone just getting into machine learning can write more efficient implementation of the algorithms than the people who have PhDs. But that's just my two cents.
A more hands-on/practical course than Dr. Ng's ML course. However, all of this code seems very repetative (and maybe that's what it is supposed to be with all of the models doing the same thing?). That being said, they went through the data preprocessing stage and gave a good machine learning template to work from for almost any project. The repetativeness may have helped with the overwhelming number of machine learning algorithms that seem to all do the same thing, just slightly differing on accuracy depending on the dataset. Does this point to using NN most of the time? It seems that they are the most handy; however, the others may be better at data collection and feature creation?
A course with a good overview of Mongo and the flask_restful library. The Docker portion of the course is very much lacking. There are much better ways of using docker-compose in a development setting (you don't have to write a single line dockerfile to pull in a mongo image to build in your docker-compose file). I found the flask_restful part helpful, but could have learned all that I did from a multi-step tutorial online. Coding standards were not a thing (showing the inexperience of the teacher). Code was copied and pasted too much (which he did say wasn't the best. But why keep doing it?). The examples weren't as complex as I would have hoped.
I have only just started this course, but it shows great promise! Thus far, it has been good. The code in the video is a little outdated, but the teacher does a really good job of pointing out where you need to update the code. The code is provided which isn't my favorite (very easy to think I understand what is happening if I don't force myself to write it all myself and make sure I understand what is happening). But I see a lot of sections that are exciting and are used in industry: testing, deploying, third-party integration, etc. The one thing that I would have liked to see is a separation of the project into frontend and backend projects so that they could be easily scaled. However, I am not sure that this cannot be easily scaled as is the way that it is written. We will have to wait and find out.