This is a repository I use to keep all the Machine Learning algorithms I've developed in one place. I have developed these algorithms either for some class I took or just for practice.
Update 26/03/2021: I've revived this repository. I've been using it mainly as a study tool, in order to implement algorithms I haven't worked with before or haven't worked with in a long time. I will most likely be adding algorithms from Stanford's CS229 class and Bishop's PRML book (back to basics!). Later on I might move on to stuff that's closer to what I actually work with (mainly Deep Learning).
Some of the code follows scikit-learn conventions, but not all of it does. In particular, none of the Ensemble methods do. Furthermore, even the ones that do follow the conventions are not guaranteed to pass scikit-learn tests.
There are some simple tests, but I can't guarantee that these algorithms are correct for every possible case. Think of them more as a reference.
Documentation will be improved with time.
- Python >= 3.8 (Haven't tested with older versions)
- Scikit-learn >= 0.24.1
I have installed these dependencies using conda, so keep in mind that other packaging systems might have different versions for scikit-learn's dependencies.
Also, if you're looking for libraries for Ensemble Methods, check out brew and DESlib