Index of modules

Type of block diagonal matrices

Module for making (co-)variance predictions

Covariance of a constant function

Covariance of linear functions with Automatic Relevance Determination

Covariance of linear functions with one hyperparameter

Module for sampling (multiple) points from the posterior distribution accounting for their covariance

Feature-rich ("fat") squared exponential covariance

Isotropic squared exponential covariance

Posterior covariances

Sub-modules for learning with derivatives.

Sub-modules for learning without derivatives.

Derivatives always require evaluation functions

Evaluation

Optimization with the GNU Scientific library (GSL)

Hyper parameters that have derivatives

Module for inducing inputs with derivatives

Evaluating inducing inputs

Derivatives of the covariance matrix of inducing inputs

Signature for evaluating inducing inputs

Evaluating single inputs

Signature for evaluating single inputs

Module for inputs with derivatives

Evaluating (multiple) inputs

Derivatives of the (cross-) covariance matrix of inputs.

Signature for evaluating multiple inputs

Representations of (sparse) derivative matrices

Kernel used for evaluation

Posterior mean for a single input

Module for making mean predictions

Posterior means for (multiple) inputs

(Untrained) model with derivative information

(Untrained) model - does not require targets

Optimization module for evidence maximization

Module for sampling single points from the posterior distribution

Signatures for learning sparse Gaussian processes with inducing inputs

Representation of indices into sparse matrices

Specification of covariance function derivatives

Specification of covariance function

Specifications of covariance functions (= kernels) and their derivatives

Statistics derived from trained models

Module for testing derivative code

Trained model with derivative information

Trained model - requires targets

Posterior variance for a single input

Posterior variances for (multiple) inputs