https://github.com/jaketae/ml-from-scratch

Machine learning algorithms implemented from scratch with NumPy
https://github.com/jaketae/ml-from-scratch

classification-algorithm clustering-algorithm from-scratch machine-learning regression-models

Last synced: 3 months ago
JSON representation

Machine learning algorithms implemented from scratch with NumPy

• Host: GitHub
• URL: https://github.com/jaketae/ml-from-scratch
• Owner: jaketae
• License: gpl-3.0
• Created: 2020-03-17T21:51:07.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-06-13T12:45:55.000Z (almost 5 years ago)
• Last Synced: 2024-11-30T19:34:02.311Z (5 months ago)
• Topics: classification-algorithm, clustering-algorithm, from-scratch, machine-learning, regression-models
• Language: Python
• Homepage:
• Size: 43 KB
• Stars: 5
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Machine Learning From Scratch

This repository contains my implementations of various machine learning algorithms from scratch. 

## Motivation

A quote from Richard Feynman:

> What I cannot create, I do not understand.

By constructing algorithms from the ground up, one can glean meaningful insights on how machine learning actually works. 

## Algorithms

Currently, the following list of algorithms have been implemented.

- [Linear Regression](https://jaketae.github.io/study/linear-regression/)

- [Logistic Regression](https://jaketae.github.io/study/logistic-regression/)

- [Naive Bayes](https://jaketae.github.io/study/naive-bayes/)

- [K-Nearest Neighbors](https://jaketae.github.io/study/KNN/)

- [Principal Component Analysis](https://jaketae.github.io/study/pca/)

- K-Means Clustering: *Post coming soon!*

For detailed explanations on the mathematics behind each model, visit the links to my blog, where I lay out a step-by-step derivation for each algorithm. The code introduced in the blog have been modified as class-based implementations for better presentation and readability.

More documentation and models to come!