Main Content

Support Vector Machine Regression

Support vector machines for regression models

For greater accuracy on low- through medium-dimensional data sets, train a support vector machine (SVM) model using fitrsvm.

For reduced computation time on high-dimensional data sets, efficiently train a linear regression model, such as a linear SVM model, using fitrlinear.

Apps

Regression LearnerTrain regression models to predict data using supervised machine learning

Blocks

RegressionSVM PredictPredict responses using support vector machine (SVM) regression model (Since R2020b)
RegressionLinear PredictPredict responses using linear regression model (Since R2023a)

Functions

expand all

fitrsvmFit a support vector machine regression model
predictPredict responses using support vector machine regression model
fitrlinearFit linear regression model to high-dimensional data
predictPredict response of linear regression model
fitrkernelFit Gaussian kernel regression model using random feature expansion
predictPredict responses for Gaussian kernel regression model
crossvalCross-validated support vector machine regression model
partialDependenceCompute partial dependence (Since R2020b)
plotPartialDependenceCreate partial dependence plot (PDP) and individual conditional expectation (ICE) plots
limeLocal interpretable model-agnostic explanations (LIME) (Since R2020b)
shapleyShapley values (Since R2021a)

Objects

expand all

RegressionSVMSupport vector machine regression model
CompactRegressionSVMCompact support vector machine regression model
RegressionLinearLinear regression model for high-dimensional data
RegressionPartitionedLinearCross-validated linear regression model for high-dimensional data
RegressionKernelGaussian kernel regression model using random feature expansion
RegressionPartitionedKernelCross-validated kernel model for regression

Topics