KernelEmbedding¶
- class probnum.quad.kernel_embeddings.KernelEmbedding(kernel, measure)¶
Bases:
object
Integrals over kernels against integration measures.
The available kernel embeddings are:
ExpQuad
LebesgueMeasure
ExpQuad
GaussianMeasure
Matern (1d)
LebesgueMeasure
ProductMatern
LebesgueMeasure
- Parameters:
kernel (CovarianceFunction) – Instance of a kernel.
measure (IntegrationMeasure) – Instance of an integration measure.
- Raises:
ValueError – If the input dimension of the kernel does not match the input dimension of the measure.
Methods Summary
kernel_mean
(x)Kernel mean w.r.t.
Kernel integrated in both arguments against the integration measure.
Methods Documentation
- kernel_mean(x)[source]¶
Kernel mean w.r.t. its first argument against the integration measure.
- Parameters:
x (ndarray) – shape=(n_eval, input_dim) – n_eval locations where to evaluate the kernel mean.
- Returns:
shape=(n_eval,) – The kernel integrated w.r.t. its first argument, evaluated at locations
x
.- Return type:
kernel_mean