Support Vector Machine (SVM) from Scratch - Machine Learning Math & Python [Code Fix in Description]

Published: 19 January 2024
on channel: kai

2,705

IMPORTANT: As @ruima3847 pointed out, I made a mistake with the `calc_gradient` function after the 46:00 mark. Be sure to check their pinned comment or the updated GitHub repository for the correct code changes!

This video covers Support Vector Machine (SVM) classification from scratch. This will include the math, intuition, and implementation of the SVM model with Python in a Jupyter Notebook.

We will open with the Hard Margin to build intuition on SVMs before moving into the Soft Margin model. At the Soft Margin stage, we will cover the math and implementation of the decision boundary, margin, hinge loss, cost function, and finally gradient descent to achieve a minimum cost.

View Notebook on GitHub: https://github.com/kailau02/machine-l...

Portfolio: https://kailauapps.com/

LinkedIn: / dylan-kai-lau

0:00 Overview
1:04 Hard Margin SVM
13:20 Soft Margin SVM Setup
15:18 Slack Variable
19:04 Derive Cost Function
21:33 Hinge Loss Intuition
25:47 Python Decision Boundary & Cost Function
36:25 Gradient Descent Intuition
40:50 Gradient Descent Math
45:40 Python Gradient Descent

On this page of the site you can watch the video online Support Vector Machine (SVM) from Scratch - Machine Learning Math & Python [Code Fix in Description] with a duration of hours minute second in good quality, which was uploaded by the user kai 19 January 2024, share the link with friends and acquaintances, this video has already been watched 2,705 times on youtube and it was liked by 95 viewers. Enjoy your viewing!

13,912

323