Docs Latest Docs Stable BuildStatus Coverage Status

A Julia package for Augmented and Normal Gaussian Processes.



AugmentedGaussianProcesses is a registered package and is symply installed by running

pkg> add AugmentedGaussianProcesses

Basic example

Here is a simple example to start right away :

using AugmentedGaussianProcesses
model = SVGP(compose(SqExponentialKernel(), ScaleTransform(1.0)), LogisticLikelihood(), AnalyticVI(), inducingpoints(KmeansAlg(50), X_train))
train!(model, X_train, y_train; iterations=100)
y_pred = predict_y(model, X_test)

A general comparison between this package is done on Julia GP Package Comparison.


AugmentedGaussianProcesses.jl is licensed under the MIT "Expat" license; see LICENSE for the full license text.