data_models.kernels.api

data_models.kernels.api

Classes

Name Description
Kernel
AdditiveKernel
MultiplicativeKernel
ScaleKernel
MaternKernel
RBFKernel
LinearKernel
PolynomialKernel
HammingDistanceKernel
TanimotoKernel
WassersteinKernel Kernel based on the Wasserstein distance.

Kernel

data_models.kernels.api.Kernel()

AdditiveKernel

data_models.kernels.api.AdditiveKernel()

MultiplicativeKernel

data_models.kernels.api.MultiplicativeKernel()

ScaleKernel

data_models.kernels.api.ScaleKernel()

MaternKernel

data_models.kernels.api.MaternKernel()

RBFKernel

data_models.kernels.api.RBFKernel()

LinearKernel

data_models.kernels.api.LinearKernel()

PolynomialKernel

data_models.kernels.api.PolynomialKernel()

HammingDistanceKernel

data_models.kernels.api.HammingDistanceKernel()

TanimotoKernel

data_models.kernels.api.TanimotoKernel()

WassersteinKernel

data_models.kernels.api.WassersteinKernel()

Kernel based on the Wasserstein distance.

It only works for 1D data that is monotonically increasing, as it is just calculating the integral of the absolute difference between two shapes. Only when both shapes are monotonically increasing, this integral is also a Wasserstein distance (https://arxiv.org/abs/2002.01878).

The shape are assumed to be discretized as a set of points. Make sure that the discretization is fine enough to capture the shape of the data.

Attributes

Name Type Description
squared bool If True, the squared exponential Wasserstein distance is used. Note that the squared exponential Wasserstein distance kernel is not positive definite for all lengthscales. For this reason, as default the absolute exponential Wasserstein distance is used.
lengthscale_prior Optional[AnyPrior] Prior for the lengthscale of the kernel.