Support Vector Machines

Support vector machines for binary or multiclass classification

Start Here

To train a binary support vector machine (SVM) classifier, use fitcsvm. To train an series of binary SVM classifiers using an error-correcting output code (ECOC) multiclass model, build an SVM template using templateSVM, and then pass it and the training data to fitcecoc. After training either type of classifier, label new observations or estimate posterior probabilities by passing the model and predictor data to predict.

Functions

fitcsvm Train binary support vector machine classifier
fitSVMPosterior Fit posterior probabilities
predict Predict labels for support vector machine classifiers
fitcecoc Fit multiclass models for support vector machines or other classifiers
templateSVM Support vector machine template
predict Predict labels for error-correcting output code multiclass classifiers

Classes

CompactClassificationSVM Compact support vector machine for binary classification
ClassificationSVM Support vector machine for binary classification
CompactClassificationECOC Compact multiclass model for support vector machines or other classifiers
ClassificationECOC Multiclass model for support vector machines or other classifiers
ClassificationPartitionedECOC Cross-validated multiclass model for support vector machines or other classifiers
Was this topic helpful?