SAGA-GIS Module Library Documentation (v2.1.3)

Module SVM Classification

Support Vector Machine (SVM) based classification for grids.
Reference:
Chang, C.-C. / Lin, C.-J. (2011): A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, vol.2/3, p.1-27. LIBSVM Homepage.

Parameters

 NameTypeIdentifierDescriptionConstraints
InputGridsGrid list (input)GRIDS--
Training AreasShapes (input)ROI--
OutputClassificationGrid (output)CLASSES--
OptionsScalingChoiceSCALING-Available Choices:
[0] none
[1] normalize (0-1)
[2] standardize
Default: 2
Verbose MessagesBooleanMESSAGE-Default: 0
Model SourceChoiceMODEL_SRC-Available Choices:
[0] create from training areas
[1] restore from file
Default: 0
Restore Model from FileFile pathMODEL_LOAD--
Class IdentifierTable fieldROI_ID--
Store Model to FileFile pathMODEL_SAVE--
SVM TypeChoiceSVM_TYPE-Available Choices:
[0] C-SVC
[1] nu-SVC
[2] one-class SVM
[3] epsilon-SVR
[4] nu-SVR
Default: 0
Kernel TypeChoiceKERNEL_TYPElinear: u'*v polynomial: (gamma*u'*v + coef0)^degree radial basis function: exp(-gamma*|u-v|^2) sigmoid: tanh(gamma*u'*v + coef0)Available Choices:
[0] linear
[1] polynomial
[2] radial basis function
[3] sigmoid
Default: 2
DegreeIntegerDEGREEdegree in kernel functionDefault: 3
GammaFloating pointGAMMAgamma in kernel functionDefault: 0.000000
coef0Floating pointCOEF0coef0 in kernel functionDefault: 0.000000
CFloating pointCOSTparameter C (cost) of C-SVC, epsilon-SVR, and nu-SVRDefault: 1.000000
nu-SVRFloating pointNUparameter nu of nu-SVC, one-class SVM, and nu-SVRDefault: 0.500000
SVR EpsilonFloating pointEPS_SVRepsilon in loss function of epsilon-SVRDefault: 0.100000
Cache SizeFloating pointCACHE_SIZEcache memory size in MBDefault: 100.000000
EpsilonFloating pointEPStolerance of termination criterionDefault: 0.001000
ShrinkingBooleanSHRINKINGwhether to use the shrinking heuristicsDefault: 0
Probability EstimatesBooleanPROBABILITYwhether to train a SVC or SVR model for probability estimatesDefault: 0
Cross ValidationIntegerCROSSVALn-fold cross validation: n must > 1Minimum: 1
Default: 1

Command-line

Usage: saga_cmd imagery_svm 0 -GRIDS <str> [-CLASSES <str>] [-SCALING <str>] [-MESSAGE <str>] [-MODEL_SRC <str>] [-MODEL_LOAD <str>] -ROI <str> [-ROI_ID <str>] [-MODEL_SAVE <str>] [-SVM_TYPE <str>] [-KERNEL_TYPE <str>] [-DEGREE <num>] [-GAMMA <str>] [-COEF0 <str>] [-COST <str>] [-NU <str>] [-EPS_SVR <str>] [-CACHE_SIZE <str>] [-EPS <str>] [-SHRINKING <str>] [-PROBABILITY <str>] [-CROSSVAL <num>]
  -GRIDS:<str>      	Grids
	Grid list (input)
  -CLASSES:<str>    	Classification
	Grid (output)
  -SCALING:<str>    	Scaling
	Choice
	Available Choices:
	[0] none
	[1] normalize (0-1)
	[2] standardize
	Default: 2
  -MESSAGE:<str>    	Verbose Messages
	Boolean
	Default: 0
  -MODEL_SRC:<str>  	Model Source
	Choice
	Available Choices:
	[0] create from training areas
	[1] restore from file
	Default: 0
  -MODEL_LOAD:<str> 	Restore Model from File
	File path
  -ROI:<str>        	Training Areas
	Shapes (input)
  -ROI_ID:<str>     	Class Identifier
	Table field
  -MODEL_SAVE:<str> 	Store Model to File
	File path
  -SVM_TYPE:<str>   	SVM Type
	Choice
	Available Choices:
	[0] C-SVC
	[1] nu-SVC
	[2] one-class SVM
	[3] epsilon-SVR
	[4] nu-SVR
	Default: 0
  -KERNEL_TYPE:<str>	Kernel Type
	Choice
	Available Choices:
	[0] linear
	[1] polynomial
	[2] radial basis function
	[3] sigmoid
	Default: 2
  -DEGREE:<num>     	Degree
	Integer
	Default: 3
  -GAMMA:<str>      	Gamma
	Floating point
	Default: 0.000000
  -COEF0:<str>      	coef0
	Floating point
	Default: 0.000000
  -COST:<str>       	C
	Floating point
	Default: 1.000000
  -NU:<str>         	nu-SVR
	Floating point
	Default: 0.500000
  -EPS_SVR:<str>    	SVR Epsilon
	Floating point
	Default: 0.100000
  -CACHE_SIZE:<str> 	Cache Size
	Floating point
	Default: 100.000000
  -EPS:<str>        	Epsilon
	Floating point
	Default: 0.001000
  -SHRINKING:<str>  	Shrinking
	Boolean
	Default: 0
  -PROBABILITY:<str>	Probability Estimates
	Boolean
	Default: 0
  -CROSSVAL:<num>   	Cross Validation
	Integer
	Minimum: 1
	Default: 1