Contents

Description of demo_multilabel_logistics.m

Demonstrates use of independent logistic regression classifiers for each candidate class for multilabel classification

clear all
close all
generateData_multiLabel

usage of independent logistic regression

options = struct('nLabels',nLabels);
model = ml_multilabel_independent(Xtrain,ytrain,options);
yhatTest = model.predict(model, Xtest);
yhatTrain = model.predict(model, Xtrain);
testError = sum(ytest~=yhatTest)/length(ytest);
model.trainError = sum(ytrain~=yhatTrain)/length(ytrain);
fprintf('Averaged misclassification test error with %s is: %.3f\n',...
        model.name, testError);
linear_makeOneContourPlot(Xtrain,ytrain, model)

Averaged misclassification test error with Independent Logistic Classifiers is: 0.010

usage of independent logistic regression with L2-regularization

options = struct('nLabels',nLabels,'lambdaL2',1e-4);
model = ml_multilabel_independent(Xtrain,ytrain,options);
yhatTest = model.predict(model, Xtest);
yhatTrain = model.predict(model, Xtrain);
testError = sum(ytest~=yhatTest)/length(ytest);
model.trainError = sum(ytrain~=yhatTrain)/length(ytrain);
fprintf('Averaged misclassification test error with %s is: %.3f\n',...
        model.name, testError);
linear_makeOneContourPlot(Xtrain,ytrain, model)

Averaged misclassification test error with Independent Logistic Classifiers is: 0.020

Published with MATLAB® R2015a