/SVM-Visualizer-Web-App

A web application to see effect of C hyperparameter on classification boundary and marginal threshold in SVM.

Primary LanguageHTMLMIT LicenseMIT

SVM Visualizer Web App

This web app shows the effect of the hyperparameters, C and kernel, on the Support Vector Machine(SVM) algorithm.

The user is asked to generate random data. The user is then given a choice of values to be used as C and kernel. Based on the user's choice, the SVM decision boundary and marginal threshold are shown. Change the values of C and the kernel to see the plot change.

ADDED :

--> Feature to see the 3D view of non-linearly separable data points, as projected by the Gaussian radial bias function.

Frontend :

  1. HTML, CSS, Bootstrap, Javascript

  2. Jinja

Backend :

  1. Python (Flask)

  2. ML model used : SVC (Support Vector Classifier)

Screenshots of the working app

image

image

image

image

image

image

License :

MIT

  • Still in development phase
  • Contributions are welcome

View the web app at : https://svm-visualizer.herokuapp.com/