This project contains basic implementation of a few fundamental machine learning algorithms in numpy. The goal of the project is to practice numpy and python as well as to understand these classical algorithms both theoretically and practically through the implementation. While some care was taken into the implementation, these are not optimized.
A few classical algorithms are implemented:
- Linear regression (L2 regularization)
- Non-parametric meta model to estimate prediction intervals using bootstrapping
- Locally weighted regression (LOWESS)
- Kernel regression
- Logistic regression (L2 regularization)
- K-NN regressor and classifier (Euclidean distance)
- Decision tree classifier (entropy for splits)
- Random forest classifier (entropy for splits)
- K-Means clustering algorithm
- Gaussian Mixture model
- PCA dimension reduction
Just clone this repository:
git clone https://github.com/clabrugere/numpy-basics.git
- python 3.7
- numpy
- matplotlib
- sklearn
This project is licensed under the MIT License - see the LICENSE.md file for details