Speed: Collects tests regarding speed of different methods and strategies. More...

Detailed Description

Speed: Collects tests regarding speed of different methods and strategies.

Author: Daniel Wirtz

Date: 2013-08-30

New in 0.7:: (Daniel Wirtz, 2013-08-30) Added this class.

This class is part of the framework

KerMor - Model Order Reduction using Kernels

Homepage http://www.morepas.org/software/index.html
Documentation http://www.morepas.org/software/kermor/index.html
License KerMor license conditions

Definition at line 18 of file Speed.m.

Static Public Member Functions
static function [ colvec < double > times , double e ] =	KernelExpCustomBaseEval (kernels.KernelExpansion kexp,integer numpts)
	Tests the evaluation speed and determines the evaluation error of a kernel expansion and this expansion using the default direct translate base. More...

static function pt =	TryCatch (loopsize)
	TryCatch: Demonstrate how slow try-catch blocks are. More...

static function pt =	BinaryvsMatSave (double fill,integer n,integer m)
	BinaryvsMatSaveSpeed: Tests the speeds and storage size required for storing double matrices. More...

static function	FindVecInMatrix (integer n,integer m)
	Created for test purposes of finding a vector in a matrix. More...

static function	AffParamMatrix (integer n)
	Tests the evaluation speed of the general.AffParamMatrix compared to loop-type evaluations of affine decompositions. More...

static function res =	GaussMexSpeedTest1Arg (sx, sy, iter)
	Tests the speed of the c implementations of evaluate for gaussians. More...

static function res =	GaussMexSpeedTest2Arg (sx, sy1, sy2, iter)
	Tests the speed of the c implementations of evaluate for gaussians. More...

Member Function Documentation

function testing.Speed.AffParamMatrix ( integer n )

static

Tests the evaluation speed of the general.AffParamMatrix compared to loop-type evaluations of affine decompositions.

Author: Daniel Wirtz

Date: 2011-10-25

New in 0.5:: (Daniel Wirtz, 2011-10-25) Added this function.

Parameters

n	The number of test runs to perform. Default: 5000

Definition at line 300 of file Speed.m.

References k, and t.

function pt = testing.Speed.BinaryvsMatSave	(	double	fill,
		integer	n,
		integer	m
	)

static

BinaryvsMatSaveSpeed: Tests the speeds and storage size required for storing double matrices.

Author: Daniel Wirtz

Date: 2012-07-09

New in 0.6:: (Daniel Wirtz, 2012-07-09) Added this function.

This class is part of the framework

KerMor - Model Order Reduction using Kernels

Homepage http://www.morepas.org/software/index.html
Documentation http://www.morepas.org/software/kermor/index.html
License KerMor license conditions

Parameters

fill	The percentage in [0,1] of which the matrix should consist of non-zero values Default: .5
n	Row number Default: 2000
m	Row number Default: 1000

Generated fields of pt:

Definition at line 172 of file Speed.m.

References PrintTable.addRow(), KerMor.App(), k, and t.

Here is the call graph for this function:

function testing.Speed.FindVecInMatrix	(	integer	n,
		integer	m
	)

static

Created for test purposes of finding a vector in a matrix.

Author: Daniel Wirtz

Date: 2011-04-01

New in 0.5:: (Daniel Wirtz, 2011-04-01) Added this function.

Parameters

n	Number of matrix rows Default: 500
m	Number of matrix columns Default: 6000

Definition at line 256 of file Speed.m.

References t.

function res = testing.Speed.GaussMexSpeedTest1Arg	(	sx,
		sy,
		iter
	)

static

Tests the speed of the c implementations of evaluate for gaussians.

One-Argument test (self-evaluation)

Note: You need to compile/mex the files in the "+kernels/@GaussKernel" folder for this to work.

Definition at line 445 of file Speed.m.

References k, and t.

function res = testing.Speed.GaussMexSpeedTest2Arg	(	sx,
		sy1,
		sy2,
		iter
	)

static

Tests the speed of the c implementations of evaluate for gaussians.

Two-Argument test.

Note: You need to compile/mex the files in the "+kernels/@GaussKernel" folder for this to work.

Definition at line 504 of file Speed.m.

References k, and t.

function [ colvec< double > times , double e ] = testing.Speed.KernelExpCustomBaseEval	(	kernels.KernelExpansion	kexp,
		integer	numpts
	)

static

Tests the evaluation speed and determines the evaluation error of a kernel expansion and this expansion using the default direct translate base.

Must have a custom base set (HasCustomBase = true)

Parameters

kexp	A kernel expansion
numpts	The number of random points at which to evaluate

Default:: 1000

Return values

times	A \(2\times 1\) vector containing the evaluation times using the custom and direct base in the first and second entry, respectively.
e	The maximum pointwise absolute and relative \(L^2\)-errors over all runs