SAGA 9.4.2 | Tool Library Documentation

Artificial Neural Network Classification


Description

Integration of the OpenCV Machine Learning library for Artificial Neural Network classification of gridded features.


References


Parameters

 NameTypeIdentifierDescriptionConstraints
InputFeaturesgrid list, inputFEATURES--
Training Samplestable, inputTRAIN_SAMPLESProvide a class identifier in the first field followed by sample data corresponding to the input feature grids.-
Training Areasshapes, inputTRAIN_AREAS--
OutputClassificationgrid, outputCLASSES--
Look-up Tabletable, output, optionalCLASSES_LUTA reference list of the grid values that have been assigned to the training classes.-
OptionsNormalizebooleanNORMALIZE-Default: 0
Colors from Featuresboolean, GUIRGB_COLORSUse the first three features in list to obtain blue, green, red components for class colour in look-up table.Default: 0
Grid Systemgrid systemGRID_SYSTEM--
TrainingchoiceMODEL_TRAIN-Available Choices: [0] training areas [1] training samples [2] load from file Default: 0
Class Identifiertable fieldTRAIN_CLASS--
Buffer Sizefloating point numberTRAIN_BUFFERFor non-polygon type training areas, creates a buffer with a diameter of specified size.Minimum: 0.000000 Default: 1.000000
Load Modelfile pathMODEL_LOADUse a model previously stored to file.-
Save Modelfile pathMODEL_SAVEStores model to file to be used for subsequent classifications instead of training areas.-
Number of Layersinteger numberANN_LAYERSYou can specify the number of layers in the network (not including input and output layer).Minimum: 1 Default: 3
Number of Neuronsinteger numberANN_NEURONSYou can specify the number of neurons in each layer of the network.Minimum: 1 Default: 5
Maximum Number of Iterationsinteger numberANN_MAXITER-Minimum: 1 Default: 300
Error Change (Epsilon)floating point numberANN_EPSILONTermination criteria of the training algorithm. You can specify how much the error could change between the iterations to make the algorithm continue (epsilon).Minimum: 0.000000 Default: 0.000000
Activation FunctionchoiceANN_ACTIVATION-Available Choices: [0] Identity [1] Sigmoid [2] Gaussian Default: 1
Function's Alphafloating point numberANN_ACT_ALPHA-Default: 1.000000
Function's Betafloating point numberANN_ACT_BETA-Default: 1.000000
Training MethodchoiceANN_PROPAGATION-Available Choices: [0] resilient propagation [1] back propagation Default: 1
Initial Update Valuefloating point numberANN_RP_DW0-Default: 0.000000
Increase Factorfloating point numberANN_RP_DW_PLUS-Minimum: 1.010000 Default: 1.200000
Decrease Factorfloating point numberANN_RP_DW_MINUS-Minimum: 0.010000 Maximum: 0.990000 Default: 0.500000
Lower Value Update Limitfloating point numberANN_RP_DW_MIN-Minimum: 0.010000 Default: 0.100000
Upper Value Update Limitfloating point numberANN_RP_DW_MAX-Minimum: 1.010000 Default: 1.100000
Weight Gradient Termfloating point numberANN_BP_DW-Default: 0.100000
Moment Termfloating point numberANN_BP_MOMENT-Default: 0.100000

Command Line


Usage: saga_cmd imagery_opencv 11 [-FEATURES ] [-NORMALIZE ] [-CLASSES ] [-CLASSES_LUT ] [-MODEL_TRAIN ] [-TRAIN_SAMPLES ] [-TRAIN_AREAS ] [-TRAIN_CLASS ] [-TRAIN_BUFFER ] [-MODEL_LOAD ] [-MODEL_SAVE ] [-ANN_LAYERS ] [-ANN_NEURONS ] [-ANN_MAXITER ] [-ANN_EPSILON ] [-ANN_ACTIVATION ] [-ANN_ACT_ALPHA ] [-ANN_ACT_BETA ] [-ANN_PROPAGATION ] [-ANN_RP_DW0 ] [-ANN_RP_DW_PLUS ] [-ANN_RP_DW_MINUS ] [-ANN_RP_DW_MIN ] [-ANN_RP_DW_MAX ] [-ANN_BP_DW ] [-ANN_BP_MOMENT ]
  -FEATURES:          	Features
	grid list, input
  -NORMALIZE:         	Normalize
	boolean
	Default: 0
  -CLASSES:           	Classification
	grid, output
  -CLASSES_LUT:       	Look-up Table
	table, output, optional
  -MODEL_TRAIN:       	Training
	choice
	Available Choices:
	[0] training areas
	[1] training samples
	[2] load from file
	Default: 0
  -TRAIN_SAMPLES:     	Training Samples
	table, input
  -TRAIN_AREAS:       	Training Areas
	shapes, input
  -TRAIN_CLASS:       	Class Identifier
	table field
  -TRAIN_BUFFER:   	Buffer Size
	floating point number
	Minimum: 0.000000
	Default: 1.000000
  -MODEL_LOAD:        	Load Model
	file path
  -MODEL_SAVE:        	Save Model
	file path
  -ANN_LAYERS:        	Number of Layers
	integer number
	Minimum: 1
	Default: 3
  -ANN_NEURONS:       	Number of Neurons
	integer number
	Minimum: 1
	Default: 5
  -ANN_MAXITER:       	Maximum Number of Iterations
	integer number
	Minimum: 1
	Default: 300
  -ANN_EPSILON:    	Error Change (Epsilon)
	floating point number
	Minimum: 0.000000
	Default: 0.000000
  -ANN_ACTIVATION:    	Activation Function
	choice
	Available Choices:
	[0] Identity
	[1] Sigmoid
	[2] Gaussian
	Default: 1
  -ANN_ACT_ALPHA:  	Function's Alpha
	floating point number
	Default: 1.000000
  -ANN_ACT_BETA:   	Function's Beta
	floating point number
	Default: 1.000000
  -ANN_PROPAGATION:   	Training Method
	choice
	Available Choices:
	[0] resilient propagation
	[1] back propagation
	Default: 1
  -ANN_RP_DW0:     	Initial Update Value
	floating point number
	Default: 0.000000
  -ANN_RP_DW_PLUS: 	Increase Factor
	floating point number
	Minimum: 1.010000
	Default: 1.200000
  -ANN_RP_DW_MINUS:	Decrease Factor
	floating point number
	Minimum: 0.010000
	Maximum: 0.990000
	Default: 0.500000
  -ANN_RP_DW_MIN:  	Lower Value Update Limit
	floating point number
	Minimum: 0.010000
	Default: 0.100000
  -ANN_RP_DW_MAX:  	Upper Value Update Limit
	floating point number
	Minimum: 1.010000
	Default: 1.100000
  -ANN_BP_DW:      	Weight Gradient Term
	floating point number
	Default: 0.100000
  -ANN_BP_MOMENT:  	Moment Term
	floating point number
	Default: 0.100000