smac.epm.gaussian_process.kernels

Functions

get_conditional_hyperparameters(X[, Y])

Returns conditional hyperparameters.

Classes

ConstantKernel([constant_value, ...])
HammingKernel([length_scale, ...])
MagicMixin()
Matern([length_scale, length_scale_bounds, ...])
Product(k1, k2[, operate_on, has_conditions])
RBF([length_scale, length_scale_bounds, ...])
Sum(k1, k2[, operate_on, has_conditions])
WhiteKernel([noise_level, ...])

class smac.epm.gaussian_process.kernels.ConstantKernel(constant_value=1.0, constant_value_bounds=(1e-05, 100000.0), operate_on=None, prior=None, has_conditions=False)

Bases: smac.epm.gaussian_process.kernels.MagicMixin, sklearn.gaussian_process.kernels.ConstantKernel

class smac.epm.gaussian_process.kernels.HammingKernel(length_scale=1.0, length_scale_bounds=(1e-05, 100000.0), operate_on=None, prior=None, has_conditions=False)

Bases: smac.epm.gaussian_process.kernels.MagicMixin, sklearn.gaussian_process.kernels.StationaryKernelMixin, sklearn.gaussian_process.kernels.NormalizedKernelMixin, sklearn.gaussian_process.kernels.Kernel

property hyperparameter_length_scale: sklearn.gaussian_process.kernels.Hyperparameter

Hyperparameter of the length scale.

Return type

Hyperparameter

class smac.epm.gaussian_process.kernels.Matern(length_scale=1.0, length_scale_bounds=(1e-05, 100000.0), nu=1.5, operate_on=None, prior=None, has_conditions=False)

Bases: smac.epm.gaussian_process.kernels.MagicMixin, sklearn.gaussian_process.kernels.Matern

class smac.epm.gaussian_process.kernels.Product(k1, k2, operate_on=None, has_conditions=False)

Bases: smac.epm.gaussian_process.kernels.MagicMixin, sklearn.gaussian_process.kernels.Product

class smac.epm.gaussian_process.kernels.RBF(length_scale=1.0, length_scale_bounds=(1e-05, 100000.0), operate_on=None, prior=None, has_conditions=False)

Bases: smac.epm.gaussian_process.kernels.MagicMixin, sklearn.gaussian_process.kernels.RBF

class smac.epm.gaussian_process.kernels.Sum(k1, k2, operate_on=None, has_conditions=False)

Bases: smac.epm.gaussian_process.kernels.MagicMixin, sklearn.gaussian_process.kernels.Sum

class smac.epm.gaussian_process.kernels.WhiteKernel(noise_level=1.0, noise_level_bounds=(1e-05, 100000.0), operate_on=None, prior=None, has_conditions=False)

Bases: smac.epm.gaussian_process.kernels.MagicMixin, sklearn.gaussian_process.kernels.WhiteKernel

smac.epm.gaussian_process.kernels.get_conditional_hyperparameters(X, Y=None)

Returns conditional hyperparameters.

Return type

ndarray

Modules

smac.epm.gaussian_process.kernels.boing