Back to Course
Machine Learning: The Engineering Approach
Module 10 of 13
10. Support Vector Machines
1. The Margin
SVMs don't just find a line; they find the widest street between classes.
- Support Vectors: The data points on the edge of the street.
- Hard vs Soft Margin: Allowing some errors to get a wider street.
2. The Kernel Trick
How do you separate data that is not linearly separable (like a circle)?
Project it into higher dimensions!
model = SVC(kernel='rbf')