saealib.MultiquadricKernel

class saealib.MultiquadricKernel(*, length_scale=None)

ベースクラス: RBFKernel

Multiquadric radial basis function kernel.

This conditionally positive definite kernel has order 0 and requires a constant polynomial term for a well-posed system. Based on [16].

パラメータ:

length_scale (float or None) -- Shape parameter (denoted gamma in the paper), in the same distance units as the input data. None resolves to the median pairwise training-point distance at fit time.

Methods

__init__
evaluate	Evaluate the multiquadric kernel sqrt(r**2 + length_scale**2).
pairwise	Evaluate the kernel between every row of x1 and x2.
resolve	Resolve length_scale from training data if unset.
validate	Validate length_scale.

Method Details

MultiquadricKernel.__init__(*, length_scale=None)

パラメータ:

length_scale (float | None)

戻り値の型:

None

MultiquadricKernel.evaluate(r)

Evaluate the multiquadric kernel sqrt(r**2 + length_scale**2).

パラメータ:

r (ndarray)

戻り値の型:

ndarray

MultiquadricKernel.pairwise(x1, x2)

Evaluate the kernel between every row of x1 and x2.

パラメータ:

• x1 (ndarray)

• x2 (ndarray)

戻り値の型:

ndarray

MultiquadricKernel.resolve(train_x)

Resolve length_scale from training data if unset.

パラメータ:

train_x (ndarray)

戻り値の型:

Self

MultiquadricKernel.validate()

Validate length_scale.

戻り値の型:

None