Scroll to navigation

OTBCLI_TRAINIMAGESCLASSIFIER(1) User Commands OTBCLI_TRAINIMAGESCLASSIFIER(1)

NAME

otbcli_TrainImagesClassifier - OTB TrainImagesClassifier application

DESCRIPTION

This is the TrainImagesClassifier application, version 5.2.0 Train a classifier from multiple pairs of images and training vector data.

Complete documentation: http://www.orfeo-toolbox.org/Applications/TrainImagesClassifier.html

Parameters:

<boolean> Report progress


-io.il <string list> Input Image List (mandatory)
-io.vd <string list> Input Vector Data List (mandatory)

<string> Input XML image statistics file (optional, off by default)
<string> Output confusion matrix (optional, off by default)


-io.out <string> Output model (mandatory)

<string> DEM directory (optional, off by default)
<string> Geoid File (optional, off by default)
<float> Default elevation (mandatory, default value is 0)
<int32> Maximum training sample size per class (mandatory, default value is 1000)
<int32> Maximum validation sample size per class (mandatory, default value is 1000)
<int32> Bound sample number by minimum (mandatory, default value is 1)
<boolean> On edge pixel inclusion (optional, off by default)
<float> Training and validation sample ratio (mandatory, default value is 0.5)
<string> Name of the discrimination field (mandatory, default value is Class)
<string> Classifier to use for the training [boost/dt/gbt/ann/bayes/rf/knn] (mandatory, default value is boost)
<string> Boost Type [discrete/real/logit/gentle] (mandatory, default value is real)
<int32> Weak count (mandatory, default value is 100)
<float> Weight Trim Rate (mandatory, default value is 0.95)
<int32> Maximum depth of the tree (mandatory, default value is 1)
<int32> Maximum depth of the tree (mandatory, default value is 65535)
<int32> Minimum number of samples in each node (mandatory, default value is 10)
<float> Termination criteria for regression tree (mandatory, default value is 0.01)
<int32> Cluster possible values of a categorical variable into K <= cat clusters to find a suboptimal split (mandatory, default value is 10)
<int32> K-fold cross-validations (mandatory, default value is 10)
<boolean> Set Use1seRule flag to false (optional, off by default)
<boolean> Set TruncatePrunedTree flag to false (optional, off by default)
<int32> Number of boosting algorithm iterations (mandatory, default value is 200)
<float> Regularization parameter (mandatory, default value is 0.01)
<float> Portion of the whole training set used for each algorithm iteration (mandatory, default value is 0.8)
<int32> Maximum depth of the tree (mandatory, default value is 3)
<string> Train Method Type [reg/back] (mandatory, default value is reg)
<string list> Number of neurons in each intermediate layer (mandatory)
<string> Neuron activation function type [ident/sig/gau] (mandatory, default value is sig)
<float> Alpha parameter of the activation function (mandatory, default value is 1)
<float> Beta parameter of the activation function (mandatory, default value is 1)
<float> Strength of the weight gradient term in the BACKPROP method (mandatory, default value is 0.1)
<float> Strength of the momentum term (the difference between weights on the 2 previous iterations) (mandatory, default value is 0.1)
<float> Initial value Delta_0 of update-values Delta_{ij} in RPROP method (mandatory, default value is 0.1)
<float> Update-values lower limit Delta_{min} in RPROP method (mandatory, default value is 1e-07)
<string> Termination criteria [iter/eps/all] (mandatory, default value is all)
<float> Epsilon value used in the Termination criteria (mandatory, default value is 0.01)
<int32> Maximum number of iterations used in the Termination criteria (mandatory, default value is 1000)
<int32> Maximum depth of the tree (mandatory, default value is 5)
<int32> Minimum number of samples in each node (mandatory, default value is 10)
<float> Termination Criteria for regression tree (mandatory, default value is 0)
<int32> Cluster possible values of a categorical variable into K <= cat clusters to find a suboptimal split (mandatory, default value is 10)
<int32> Size of the randomly selected subset of features at each tree node (mandatory, default value is 0)
Maximum number of trees in the forest (mandatory, default value is 100)
<float> Sufficient accuracy (OOB error) (mandatory, default value is 0.01)
<int32> Number of Neighbors (mandatory, default value is 32)
<int32> set user defined seed (optional, off by default)
<string> Load otb application from xml file (optional, off by default)

EXAMPLES

otbcli_TrainImagesClassifier -io.il QB_1_ortho.tif -io.vd VectorData_QB1.shp -io.imstat EstimateImageStatisticsQB1.xml -sample.mv 100 -sample.mt 100 -sample.vtr 0.5 -sample.edg false -sample.vfn Class -classifier libsvm -classifier.libsvm.k linear -classifier.libsvm.c 1 -classifier.libsvm.opt false -io.out svmModelQB1.txt -io.confmatout svmConfusionMatrixQB1.csv

SEE ALSO

The full documentation for otbcli_TrainImagesClassifier is maintained as a Texinfo manual. If the info and otbcli_TrainImagesClassifier programs are properly installed at your site, the command

info otbcli_TrainImagesClassifier

should give you access to the complete manual.

December 2015 otbcli_TrainImagesClassifier 5.2.0