Support vector machine

In statistical learning, support vector machines are a supervised approach for classification and regression.

Resources

• An idiot’s guide to support vector machines (SVMs), by R. Berwick

In code

scikit-learn offers several support vector machine classes for different tasks. Within the base class sklearn.svm, we have the sub-classes:

• For classification:
  • SVC
  • NuSVC
  • LinearSVC
• For regression:
  • SVR
  • NuSVR
  • LinearSVR