Main Content

Gaussian Process Regression

Gaussian process regression models (kriging)

Gaussian process regression (GPR) models are nonparametric, kernel-based probabilistic models. To train a GPR model interactively, use the Regression Learner app. For greater flexibility, train a GPR model using the fitrgp function at the command line. After training, you can predict responses for new data by passing the model and the new predictor data to the predict object function.


Regression LearnerTrain regression models to predict data using supervised machine learning


RegressionGP PredictPredict responses using Gaussian process (GP) regression model (Since R2022a)


expand all

fitrgpFit a Gaussian process regression (GPR) model
compactReduce size of machine learning model
templateGPGaussian process template (Since R2023b)
limeLocal interpretable model-agnostic explanations (LIME) (Since R2020b)
partialDependenceCompute partial dependence (Since R2020b)
permutationImportancePredictor importance by permutation (Since R2024a)
plotPartialDependenceCreate partial dependence plot (PDP) and individual conditional expectation (ICE) plots
shapleyShapley values (Since R2021a)
crossvalCross-validate machine learning model
kfoldLossLoss for cross-validated partitioned regression model
kfoldPredictPredict responses for observations in cross-validated regression model
kfoldfunCross-validate function for regression
lossRegression error for Gaussian process regression model
resubLossResubstitution regression loss
postFitStatisticsCompute post-fit statistics for the exact Gaussian process regression model
predictPredict response of Gaussian process regression model
resubPredictPredict responses for training data using trained regression model


RegressionGPGaussian process regression model
CompactRegressionGPCompact Gaussian process regression model class
RegressionPartitionedGPCross-validated Gaussian process regression (GPR) model (Since R2022b)