https://github.com/balaka-18/svm_visual_tool

A simple web app that helped students visualize the SVM algorithm according to their choice of hyperparameter setting.
https://github.com/balaka-18/svm_visual_tool

decision-boundary decision-boundary-visualizations svm svm-classifier svm-learning webapp

Last synced: 7 months ago
JSON representation

A simple web app that helped students visualize the SVM algorithm according to their choice of hyperparameter setting.

• Host: GitHub
• URL: https://github.com/balaka-18/svm_visual_tool
• Owner: BALaka-18
• License: mit
• Created: 2020-08-30T03:53:31.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2020-09-23T22:59:02.000Z (over 5 years ago)
• Last Synced: 2025-06-06T23:08:38.706Z (7 months ago)
• Topics: decision-boundary, decision-boundary-visualizations, svm, svm-classifier, svm-learning, webapp
• Language: Python
• Homepage:
• Size: 20.5 KB
• Stars: 3
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# SVM VISUAL TOOL : A VISUALIZATION TOOL TO HELP STUDENTS UNDERSTAND SUPPORT VECTOR MACHINES BETTER.

VISIT THE WEB APP HERE : [Welcome to the SVM Visual Tool(SVM VT)](https://svm-main.herokuapp.com/)

YOUTUBE VIDEO LINK : [Watch it here - Support Vector Machine Visualizer](https://www.youtube.com/watch?v=LYzz-CHEXD0)

Support Vector Machines is one of the most famous supervised learning algorithms in Data Science.

However, understanding how hyperparameters can affect the performance of this algorithm is quite tricky for a beginner.

When I had started learning SVM, I had found it difficult to imagine or picturize how the decision boundary and threshold lines change when I change the values of the hyperparameters or switch between kernels.

So, I decided to make a simple web app that helped students visualize the SVM algorithm according to their choice of hyperparameter setting.

### About the web app :

The web app consists of two pages :

1. The HOME page : The home page is basically the gateway to the visual tool. It also consists a link that redirects the user to blogs on SVM, in case the user wants to have a short read.

HOME PAGE SCREENSHOT :

![Screenshot (349)](https://user-images.githubusercontent.com/49288068/94067686-de8b7c00-fe0b-11ea-9522-14f3a09e98e2.png)

2. The main page : The SVM VISUAL TOOL

This page contains the main plot that is generated upon changing the settings below the plot on the left hand side.

THE PLOTS :

![Screenshot (350)](https://user-images.githubusercontent.com/49288068/94067700-e3503000-fe0b-11ea-8cf5-85d8449358e6.png)

THE CONTROLS AND THE SHORT NOTES TO EXPLAIN THE HYPERPARAMETERS :

![Screenshot (351)](https://user-images.githubusercontent.com/49288068/94067718-e814e400-fe0b-11ea-8a4b-c8be6819846e.png)

Tech stack :

Frontend : HTML, CSS, Bootstrap, Flask.

Backend : Dash, Python.

________________________________________________________________________________________________________________________________________________________________________________