Writing import sklearn.some_xyz_model as xyz and xyz.fit() is very convenient but learning how the code works underneath is more fun. Therefore, I challenged myself to fully understand and write popular ML algorithms from scratch.
I decided to document my data science graduate journey by actually putting into practice what I learned at grad school. This repository demonstrates raw implementations of popular ML algorithms using just linear algebra library like Numpy and PyTorch (to leverage GPU) and some helper libraries like pandas. Pre-processing of data is also done from scratch (most of the time).
PRs are not welcome, sorry!