This is a simple machine learning library that I am writing using only numpy, in order to have a better understanding of the main machine learning algorithms.
fbml stands for "François BONIFACE machine learning"
- Linear regresssion: standard, Lasso, Ridge, Elastic-Net
- Logistic regression with l1 or l2 penalty
- support vector machine for classification: runnable but not working properly. The goal was to understand how SVM work and it was achieved. I got tired of debugging, so I am leaving it as it is for now.