Abstract class to implement non-parametric processes. More...
#include <kernelregressor.hpp>
Inheritance diagram for bayesopt::KernelRegressor:
Collaboration diagram for bayesopt::KernelRegressor:
Public Member Functions | |
| KernelRegressor (size_t dim, Parameters parameters, const Dataset &data, MeanModel &mean, randEngine &eng) | |
| void | fitSurrogateModel () |
| Computes the initial surrogate model and updates the kernel parameters estimation. More... | |
| void | updateSurrogateModel () |
| Sequential update of the surrogate model by adding a new row to the Kernel matrix, more precisely, to its Cholesky decomposition. More... | |
| double | getSignalVariance () |
| size_t | nHyperParameters () |
| vectord | getHyperParameters () |
| void | setHyperParameters (const vectord &theta) |
| Public Member Functions inherited from bayesopt::NonParametricProcess | |
| NonParametricProcess (size_t dim, Parameters parameters, const Dataset &data, MeanModel &mean, randEngine &eng) | |
| virtual ProbabilityDistribution * | prediction (const vectord &query)=0 |
| Function that returns the prediction of the GP for a query point in the hypercube [0,1]. More... | |
| double | getValueAtMinimum () |
| const Dataset * | getData () |
| double | getSignalVariance () |
| Public Member Functions inherited from bayesopt::RBOptimizable | |
| virtual double | evaluate (const vectord &query)=0 |
Protected Member Functions | |
| virtual void | precomputePrediction ()=0 |
| Sets the kind of learning methodology for kernel hyperparameters. More... | |
| void | computeCorrMatrix (matrixd &corrMatrix) |
| Computes the Correlation (Kernel or Gram) matrix. | |
| matrixd | computeCorrMatrix () |
| Computes the Correlation (Kernel or Gram) matrix. | |
| matrixd | computeDerivativeCorrMatrix (int dth_index) |
| Computes the derivative of the correlation matrix with respect to the dth hyperparameter. | |
| vectord | computeCrossCorrelation (const vectord &query) |
| double | computeSelfCorrelation (const vectord &query) |
| void | computeCholeskyCorrelation () |
| Computes the Cholesky decomposition of the Correlation matrix. | |
Protected Attributes | |
| matrixd | mL |
| Cholesky decomposition of the Correlation matrix. | |
| score_type | mScoreType |
| learning_type | mLearnType |
| bool | mLearnAll |
| KernelModel | mKernel |
| Protected Attributes inherited from bayesopt::NonParametricProcess | |
| const Dataset & | mData |
| double | mSigma |
| Signal variance. | |
| size_t | dim_ |
| MeanModel & | mMean |
Private Member Functions | |
| void | addNewPointToCholesky (const vectord &correlation, double selfcorrelation) |
| Adds a new point to the Cholesky decomposition of the Correlation matrix. More... | |
Private Attributes | |
| const double | mRegularizer |
| Std of the obs. model (also used as nugget) | |
Additional Inherited Members | |
| Static Public Member Functions inherited from bayesopt::NonParametricProcess | |
| static NonParametricProcess * | create (size_t dim, Parameters parameters, const Dataset &data, MeanModel &mean, randEngine &eng) |
| Factory model generator for surrogate models. More... | |
Detailed Description
Abstract class to implement non-parametric processes.
Definition at line 46 of file kernelregressor.hpp.
Member Function Documentation
◆ precomputePrediction()
|
protectedpure virtual |
Sets the kind of learning methodology for kernel hyperparameters.
Precompute some values of the prediction that do not depends on the query.
Implemented in bayesopt::GaussianProcess, bayesopt::GaussianProcessML, bayesopt::StudentTProcessJeffreys, bayesopt::StudentTProcessNIG, and bayesopt::GaussianProcessNormal.
Referenced by fitSurrogateModel().
◆ updateSurrogateModel()
|
virtual |
Sequential update of the surrogate model by adding a new row to the Kernel matrix, more precisely, to its Cholesky decomposition.
It assumes that the kernel hyperparemeters do not change.
Implements bayesopt::NonParametricProcess.
Definition at line 52 of file kernelregressor.cpp.
References bayesopt::utils::cholesky_add_row().
The documentation for this class was generated from the following files:
