164 this =
this@kernels.KernelExpansion;
172 this.fTK= kernels.NoKernel;
173 this.fPK= kernels.NoKernel;
180 " SubKernelCombinationFun ",
" StateNablaCombinationFun ");
191 fx = this.
Ma * (this.Base \ this.getKernelVector(x, t, mu)^
t);
197 this.fTK.evaluate(t,
this.Centers.ti), ...
199 this.fPK.evaluate(mu, this.
Centers.mui));
215 error(
" Derivaties only possible for single vector/point, as already returning a matrix with derivatives at all centers. ");
218 this.fTK.evaluate(t,
this.Centers.ti),...
220 this.fPK.evaluate(mu, this.
Centers.mui));
259 this.fTK.evaluate(
this.Centers.ti,[]), ...
261 this.fPK.evaluate(this.
Centers.mui,[]));
288 this.fTK.evaluate(t,t(idx)), ...
290 this.fPK.evaluate(mu,mu(:,idx)));
336 copy = kernels.ParamTimeKernelExpansion;
339 copy =
clone@kernels.KernelExpansion(
this, copy);
342 copy.fTK= this.fTK.clone;
343 copy.fPK= this.fPK.clone;
348 clear@kernels.KernelExpansion(
this);
363 #if 0 //mtoc++: 'set.SubKernelCombinationFun'
365 if ~isa(fhandle,
" function_handle ")
367 elseif nargin(fhandle) ~= 3
377 #if 0 //mtoc++: 'get.ParamKernel'
386 #if 0 //mtoc++: 'get.TimeKernel'
395 #if 0 //mtoc++: 'set.ParamKernel'
397 if isa(value,
" kernels.BaseKernel ")
400 error("
ParamKernel must be a subclass of kernels.BaseKernel. ");
408 #if 0 //mtoc++: 'set.TimeKernel'
410 if isa(value,
" kernels.BaseKernel ")
413 error("
TimeKernel must be a subclass of kernels.BaseKernel. ");
kernels.BaseKernel ParamKernel
The Kernel to use for parameter variables.
matrix< double > Ma
The coefficient data for each dimension.
logical hasParams
Flag that indicates if param samples are present.
function K = getKernelMatrixColumn(idx,colvec< double > x,double t,colvec< double > mu)
Computes the kernel matrix for the currently set center data.
function clear()
Removes all centers and coefficients from the expansion and leaves the associated kernels untouched...
function phi = getKernelVector(colvec< double > x,double t,colvec< double > mu)
Returns the kernel vector . (for the case of a product SubKernelCombinationFun
ParamTimeKernelExpansion()
Default constructor.
function registerProps(varargin)
Call this method at any class that defines DPCM observed properties.
virtual function Nabla = getNabla(x, y)
Computes the partial derivatives with respect to each component of the first argument.
virtual function c = getGlobalLipschitz()
Returns the global lipschitz constant of this kernel.
function fx = evaluate(colvec< double > x,double t,colvec< double > mu)
Matlab's base handle class (documentation generation substitute)
rowvec ti
The time samples .
function c = getGlobalLipschitz(double t,colvec< double > mu)
Overrides the implementation in KernelExpansion.
kernels.BaseKernel fSK
The inner (state) kernel object.
rowvec Ma_norms
The norms of the coefficient matrix of the kernel expansion.
kernels.BaseKernel Kernel
The Kernel to use for system variables.
logical hasTime
Flag that indicates if time samples are present.
matrix mui
The parameter samples used computing the parent trajectories of .
StateNablaCombinationFun
The combination function for the nabla of the system/state kernel with the other kernels.
ParamTimeKernelExpansion: Kernel expansion class for time and/or parameter dependent kernels...
kernels.BaseKernel TimeKernel
The Kernel to use for time variables.
function J = getStateJacobian(colvec< double > x,double t,colvec< double > mu)
Evaluates the jacobian matrix of this function at the given point.
function K = getKernelMatrix()
Computes the kernel matrix for the currently set center data.
function setCentersFromATD(data.ApproxTrainData atd,rowvec< integer > idx)
Sets the centers according to the indices idx of the training data.
ApproxTrainData: Data class for approximation training data, containing several useful bounding box p...
struct Centers
The kernel centers used in the approximation.
SubKernelCombinationFun
The function that combines the sub (time/system/param) kernels. Must be a function handle that takes ...
function copy = clone(copy)
The interface method with returns a copy of the current class instance.
virtual function K = evaluate(matrix< double > x,matrix< double > y)
Evaluation method for the current kernel.
Base class for all KerMor Kernels.
KernelExpansion: Base class for state-space kernel expansions.