Tool NameRandom Forest Classification (ViGrA)
Tool ID9
Library IDimagery_vigra
Version1.0
Author(s)O.Conrad (c) 2013

Description

Random Forest Classification.

References

ViGrA - Vision with Generic Algorithms


Parameters

NameTypeIdentifierDescriptionConstraints
Input
Featuresgrid list, inputFEATURES
Training Areasshapes, inputTRAINING
Output
Random Forest Classificationgrid, outputCLASSES
Prediction Probabilitygrid, output, optionalPROBABILITY
Feature Probabilitiesgrid list, outputPROBABILITIES
Feature Importancestable, outputIMPORTANCES
Options
Feature ProbabilitiesbooleanBPROBABILITIESDefault: 0
Label Fieldtable fieldFIELD
Label is IdentifierbooleanLABEL_AS_IDUse training area labels as identifier in classification result, assumes all label values are integer numbers!Default: 0
Minimum Redundancy Feature SelectionbooleanDO_MRMRUse only features selected by the minimum Redundancy Maximum Relevance (mRMR) algorithmDefault: 0
Number of Featuresinteger numbermRMR_NFEATURESMinimum: 1
Default: 50
DiscretizationbooleanmRMR_DISCRETIZEuncheck this means no discretizaton (i.e. data is already integer)Default: 1
Discretization Thresholdfloating point numbermRMR_THRESHOLDa double number of the discretization threshold; set to 0 to make binarizationMinimum: 0.000000
Default: 1.000000
Selection MethodchoicemRMR_METHODAvailable Choices:
[0] Mutual Information Difference (MID)
[1] Mutual Information Quotient (MIQ)
Default: 0
Load Modelfile pathRF_IMPORT
Save Modelfile pathRF_EXPORT
Tree Countinteger numberRF_TREE_COUNTHow many trees to create?Minimum: 1
Default: 32
Samples per Treefloating point numberRF_TREE_SAMPLESSpecifies the fraction of the total number of samples used per tree for learning.Minimum: 0.000000
Maximum: 1.000000
Default: 1.000000
Sample with ReplacementbooleanRF_REPLACESample from training population with or without replacement?Default: 1
Minimum Node Split Sizeinteger numberRF_SPLIT_MIN_SIZENumber of examples required for a node to be split. Choose 1 for complete growing.Minimum: 1
Default: 1
Features per NodechoiceRF_NODE_FEATURESAvailable Choices:
[0] logarithmic
[1] square root
[2] all
Default: 1
StratificationchoiceRF_STRATIFICATIONSpecifies stratification strategy. Either none, equal amount of class samples, or proportional to fraction of class samples.Available Choices:
[0] none
[1] equal
[2] proportional
Default: 0