For greater accuracy and kernel-function choices on low- through medium-dimensional data sets, train a binary SVM model or a multiclass error-correcting output codes (ECOC) model containing SVM binary learners using the Classification Learner app. For greater flexibility, use the command-line interface to train a binary SVM model using fitcsvm or train a multiclass ECOC model composed of binary SVM learners using fitcecoc.
For reduced computation time on high-dimensional data sets, efficiently train a binary, linear classification model, such as a linear SVM model, using fitclinear or train a multiclass ECOC model composed of SVM models using fitcecoc.
For nonlinear classification with big data, train a binary, Gaussian kernel classification model using fitckernel.
| Classification Learner | Train models to classify data using supervised machine learning |
| ClassificationSVM Predict | Classify observations using support vector machine (SVM) classifier for one-class and binary classification |
Train Support Vector Machines Using Classification Learner App
Create and compare support vector machine (SVM) classifiers, and export trained models to make predictions for new data.
Support Vector Machines for Binary Classification
Perform binary classification via SVM using separating hyperplanes and kernel transformations.
Predict Class Labels Using ClassificationSVM Predict Block
This example shows how to use the ClassificationSVM Predict block for label prediction.
Signal Classification Using Wavelet-Based Features and Support Vector Machines (Wavelet Toolbox)
Wavelet Time Scattering Classification of Phonocardiogram Data (Wavelet Toolbox)