Skip to content

Files

Latest commit

 

History

History
142 lines (105 loc) · 2.32 KB

kernels.md

File metadata and controls

142 lines (105 loc) · 2.32 KB
  CurrentModule = KernelFunctions

Kernel Functions

[Base Kernels](@id base_kernels)

These are the basic kernels without any transformation of the data. They are the building blocks of KernelFunctions.

Constant Kernels

ZeroKernel
ConstantKernel
WhiteKernel
EyeKernel

Cosine Kernel

CosineKernel

Exponential Kernels

ExponentialKernel
GibbsKernel
LaplacianKernel
SqExponentialKernel
SEKernel
GaussianKernel
RBFKernel
GammaExponentialKernel

Exponentiated Kernel

ExponentiatedKernel

Fractional Brownian Motion Kernel

FBMKernel

Gabor Kernel

gaborkernel

Matérn Kernels

MaternKernel
Matern12Kernel
Matern32Kernel
Matern52Kernel

Neural Network Kernel

NeuralNetworkKernel

Periodic Kernel

PeriodicKernel
PeriodicKernel(::DataType, ::Int)

Piecewise Polynomial Kernel

PiecewisePolynomialKernel

Polynomial Kernels

LinearKernel
PolynomialKernel

Rational Kernels

RationalKernel
RationalQuadraticKernel
GammaRationalKernel

Spectral Mixture Kernels

spectral_mixture_kernel
spectral_mixture_product_kernel

Wiener Kernel

WienerKernel

Composite Kernels

The modular design of KernelFunctions uses [base kernels](@ref base_kernels) as building blocks for more complex kernels. There are a variety of composite kernels implemented, including those which [transform the inputs](@ref input_transforms) to a wrapped kernel to implement length scales, scale the variance of a kernel, and sum or multiply collections of kernels together.

TransformedKernel
∘(::Kernel, ::Transform)
ScaledKernel
KernelSum
KernelProduct
KernelTensorProduct
NormalizedKernel

Multi-output Kernels

Kernelfunctions implements multi-output kernels as scalar kernels on an extended output domain. For more details on this read [the section on inputs for multi-output GPs](@ref Inputs-for-Multiple-Outputs).

For a function f(x) \rightarrow y denote the inputs as x, x', such that we compute the covariance between output components y_{p} and y_{p'}. The total number of outputs is m.

MOKernel
IndependentMOKernel
LatentFactorMOKernel
IntrinsicCoregionMOKernel
LinearMixingModelKernel