deephyper.skopt.utils.HammingKernel

deephyper.skopt.utils.HammingKernel#

class deephyper.skopt.utils.HammingKernel(*args: Any, **kwargs: Any)[source]#

Bases: StationaryKernelMixin, NormalizedKernelMixin, Kernel

The HammingKernel is used to handle categorical inputs.

K(x_1, x_2) = exp(\sum_{j=1}^{d} -ls_j * (I(x_1j != x_2j)))

Parameters:

  • [float (* length_scale) – The length scale of the kernel. If a float, an isotropic kernel is used. If an array, an anisotropic kernel is used where each dimension of l defines the length-scale of the respective feature dimension.

  • array-like – The length scale of the kernel. If a float, an isotropic kernel is used. If an array, an anisotropic kernel is used where each dimension of l defines the length-scale of the respective feature dimension.

  • shape=[n_features – The length scale of the kernel. If a float, an isotropic kernel is used. If an array, an anisotropic kernel is used where each dimension of l defines the length-scale of the respective feature dimension.

  • ] – The length scale of the kernel. If a float, an isotropic kernel is used. If an array, an anisotropic kernel is used where each dimension of l defines the length-scale of the respective feature dimension.

  • (default)] (1e5]) – The length scale of the kernel. If a float, an isotropic kernel is used. If an array, an anisotropic kernel is used where each dimension of l defines the length-scale of the respective feature dimension.

  • [array-like (* length_scale_bounds) – The lower and upper bound on length_scale

  • [1e-5 – The lower and upper bound on length_scale

  • (default)] – The lower and upper bound on length_scale

Methods

gradient_x

Computes gradient of K(x, X_train) with respect to x

Attributes

hyperparameter_length_scale

__call__(X, Y=None, eval_gradient=False)[source]#

Return the kernel k(X, Y) and optionally its gradient.

Args: * X [array-like, shape=(n_samples_X, n_features)]

Left argument of the returned kernel k(X, Y)

  • Y [array-like, shape=(n_samples_Y, n_features) or None(default)]

    Right argument of the returned kernel k(X, Y). If None, k(X, X) if evaluated instead.

  • eval_gradient [bool, False(default)]

    Determines whether the gradient with respect to the kernel hyperparameter is determined. Only supported when Y is None.

Returns: * K [array-like, shape=(n_samples_X, n_samples_Y)]

Kernel k(X, Y)

  • K_gradient [array-like, shape=(n_samples_X, n_samples_X, n_dims)]

    The gradient of the kernel k(X, X) with respect to the hyperparameter of the kernel. Only returned when eval_gradient is True.

gradient_x(x, X_train)#

Computes gradient of K(x, X_train) with respect to x

Args: x: array-like, shape=(n_features,)

A single test point.

X_train: array-like, shape=(n_samples, n_features)

Training data used to fit the gaussian process.

Returns: gradient_x: array-like, shape=(n_samples, n_features)

Gradient of K(x, X_train) with respect to x.

Contents