Classical algorithms

Most popular machine learning algorithms written only with numpy
In our projects we often need to modify common used algorithms in
in a way that is not supported by any packages.
The algorithms are written in the form of classes, and in the additional
part we compare them with similar functions from Sklearn
All models are divided into groups. I am constantly supplementing this
project with new implementations

Classification tasks

1. Decision Tree
2. Supporting Vector Machine
3. Naive Bayes Classifier
4. K Nearest Neighbors

Data preprocessing

1. Principal Component Analysis
2. Linear Discriminant Analysis

Boosting

1. Ada Boost
2. Random Forest

Clustering

1. K-means

Regressions

1. Linear Regression