https://github.com/shilangyu/svm-from-scratch

Explanation and implementation of basic SVMs in Julia
https://github.com/shilangyu/svm-from-scratch

julia machine-learning svm

Last synced: about 1 year ago
JSON representation

Explanation and implementation of basic SVMs in Julia

• Host: GitHub
• URL: https://github.com/shilangyu/svm-from-scratch
• Owner: shilangyu
• Created: 2022-11-20T16:47:04.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2024-02-11T21:13:30.000Z (over 2 years ago)
• Last Synced: 2025-02-02T04:43:46.575Z (over 1 year ago)
• Topics: julia, machine-learning, svm
• Language: Julia
• Homepage: https://github.shilangyu.dev/SVM-from-scratch/
• Size: 2.39 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# SVM from scratch in Julia

Repo including explanation and high-level derivation of the SVM method used for classification, see [github.shilangyu.dev/SVM-from-scratch](https://github.shilangyu.dev/SVM-from-scratch). Additionally implementations of 3 different SVMs models can be found in Julia files.

## Linear soft margin SVM

![](animations/linear_soft_margin.gif)

## Non linear RBF kernel SVM

![](animations/non_linear.gif)

> _training does not look continuous because the optimization method takes harsh steps which converge quickly_