Loading…
Glossary term
A classifier that finds the boundary that best separates classes of data.
An SVM finds the boundary that separates two classes with the widest possible margin. With a kernel trick it can draw curved boundaries by implicitly working in a higher-dimensional space.
It was a leading method before deep learning for medium-sized datasets.
An SVM classifier
from sklearn.svm import SVC
svm = SVC(kernel="rbf").fit(X_train, y_train)It shows how a simple geometric idea, maximizing the margin, yields a strong classifier.