lsecTrain

Training the LSE (least-square estimate) classifier (LSEC)

Contents

Syntax

Description

cPrm=lsecTrain(DS) returns the parameters of the least-square estimate (LSE) classifier based on the given dataset DS.

cPrm=lsecTrain(DS, opt) uses the train parameters opt for training the LSEC.

cPrm=lsecTrain(DS, opt, showPlot) plots the decision boundary of the LSEC (if the feature dimensionity is 2).

[cPrm, logLike, recogRate, hitIndex]=qcTrain(DS, ...) also returns the log likelihood, recognition rate, and the hit indice of data instances in DS.

Example

DS=prData('iris');
DS.input=DS.input(3:4, :);
trainSet.input=DS.input(:, 1:2:end); trainSet.output=DS.output(:, 1:2:end);
 testSet.input=DS.input(:, 2:2:end);  testSet.output=DS.output(:, 2:2:end);
[cPrm, logLike1, R1]=lsecTrain(trainSet);
[computedClass, logLike2, R2]=lsecEval(testSet, cPrm, 1);
fprintf('Inside R-square = %g\n', R1);
fprintf('Outside R-square = %g\n', R2);

Reference to non-existent field 'class'.

Error in classifierPlot (line 119)
classNum=length(cPrm.class);

Error in classifierEval (line 257)
	classifierPlot(classifier, DS, cPrm, 'decBoundary');

Error in lsecEval (line 39)
[computedClass, logLike, recogRate, hitIndex]=classifierEval(classifier, DS, cPrm, showPlot);

Error in lsecTrain_help (line 29)
[computedClass, logLike2, R2]=lsecEval(testSet, cPrm, 1);

See Also

lsecEval.

Published with MATLAB® R2018a