Create PKLotTestbed.java

dynse/src/main/java/br/ufpr/dynse/testbed/PKLotTestbed.java

@@ -0,0 +1,116 @@
+/*    
+*    NebraskaWeatherTestBed.java 
+*    Copyright (C) 2017 Universidade Federal do Paraná, Curitiba, Paraná, Brasil
+*    @Author Paulo Ricardo Lisboa de Almeida ([email protected])
+*    This program is free software: you can redistribute it and/or modify
+*    it under the terms of the GNU General Public License as published by
+*    the Free Software Foundation, either version 3 of the License, or
+*    (at your option) any later version.
+*    
+*    This program is distributed in the hope that it will be useful,
+*    but WITHOUT ANY WARRANTY; without even the implied warranty of
+*    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+*    GNU General Public License for more details.
+*    
+*    You should have received a copy of the GNU General Public License
+*    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+*/
+package br.ufpr.dynse.testbed;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import br.ufpr.dynse.classifier.factory.AbstractDynseFactory;
+import br.ufpr.dynse.classifier.factory.RealConceptDriftDynseFactory;
+import br.ufpr.dynse.core.StreamDynse;
+import br.ufpr.dynse.core.UFPRLearningCurve;
+import br.ufpr.dynse.evaluation.EvaluateInterleavedChunksUFPR;
+import br.ufpr.dynse.evaluation.EvaluatePrequentialSingleTrainPKLot;
+import br.ufpr.dynse.evaluation.EvaluateSingleTrain;
+import br.ufpr.dynse.generator.StaggerDriftGenerator;
+import br.ufpr.dynse.util.PKLotStreamer;
+import br.ufpr.dynse.util.UFPRLearningCurveUtils;
+import moa.classifiers.meta.LeveragingBag;
+import moa.classifiers.trees.HoeffdingTree;
+import moa.streams.ArffFileStream;
+import moa.tasks.StandardTaskMonitor;
+import moa.tasks.TaskMonitor;
+
+public class PKLotTestbed implements MultipleExecutionsTestbed{
+
+	private static final int NUM_SAMPLES_EACH_BATCH = 30;
+	private static final int NUM_SAMPLES_TRAIN_CLASSIFIER = NUM_SAMPLES_EACH_BATCH*3;//accumulate 3 bathes (V=3)
+
+	private static final String PATH_DATASET = "/home/granza/Granza/PIC/PKLot-ConceptDrift";
+
+
+	private final AbstractDynseFactory dynseFactory = new RealConceptDriftDynseFactory();
+
+	private UFPRLearningCurveUtils ufprLearningCurveUtils = new UFPRLearningCurveUtils();	
+
+	@Override
+	public void executeTests(int numExec) throws Exception{
+
+		this.executeTestsSingleTrainPrequential(numExec);
+		this.executeTestsSingleTrainHeldOut(numExec);
+
+	}
+
+
+	public void executeTestsSingleTrainPrequential(int numExec) throws Exception{
+		List<UFPRLearningCurve> learningCurves = new ArrayList<UFPRLearningCurve>(numExec);
+		System.out.println("Testes prequenciais da PKLot");
+		for(int i =0;i < numExec; i++){
+			System.out.println("Teste "+(i+1)+":");
+			TaskMonitor monitor = new StandardTaskMonitor();
+			EvaluatePrequentialSingleTrainPKLot evaluator = new EvaluatePrequentialSingleTrainPKLot();
+
+			ArffFileStream stream = new ArffFileStream();
+			PKLotStreamer file = new PKLotStreamer();
+			stream.arffFileOption.setValue(file.prepareStream(PATH_DATASET));
+
+			evaluator.learnerOption.setCurrentObject(new HoeffdingTree());
+
+			evaluator.streamOption.setCurrentObject(stream);
+			evaluator.trainSizeOption.setValue((int)(300));
+			evaluator.sampleFrequencyOption.setValue(100);
+			evaluator.widthOption.setValue(1000);
+			evaluator.dumpFileOption.setValue("/home/granza/Área de Trabalho/teste.csv");
+			evaluator.prepareForUse();
+			UFPRLearningCurve lc = (UFPRLearningCurve)evaluator.doTask(monitor, null);
+			learningCurves.add(lc);
+		}
+		UFPRLearningCurve avgResult = ufprLearningCurveUtils.averageResults(learningCurves);
+
+		System.out.println(ufprLearningCurveUtils.strMainStatisticsMatlab(avgResult));
+	}
+	public void executeTestsSingleTrainHeldOut(int numExec) throws Exception{
+		List<UFPRLearningCurve> learningCurves = new ArrayList<UFPRLearningCurve>(numExec);
+		System.out.println("Testes HeldOut da PKLot");
+		for(int i =0;i < numExec; i++){
+			System.out.println("Teste "+(i+1)+":");
+			TaskMonitor monitor = new StandardTaskMonitor();
+			EvaluateSingleTrain evaluator = new EvaluateSingleTrain();
+
+			ArffFileStream stream = new ArffFileStream();
+			PKLotStreamer file = new PKLotStreamer();
+			stream.arffFileOption.setValue(file.prepareStream(PATH_DATASET));
+
+			evaluator.learnerOption.setCurrentObject(new HoeffdingTree());
+
+			evaluator.streamOption.setCurrentObject(stream);
+			evaluator.trainSizeOption.setValue((int)(300));
+			evaluator.testSizeOption.setValue(0);
+			evaluator.sampleFrequencyOption.setValue(100);
+			evaluator.dumpFileOption.setValue("/home/granza/Área de Trabalho/teste.csv");
+			evaluator.prepareForUse();
+			UFPRLearningCurve lc = (UFPRLearningCurve)evaluator.doTask(monitor, null);
+			learningCurves.add(lc);
+		}
+		UFPRLearningCurve avgResult = ufprLearningCurveUtils.averageResults(learningCurves);
+
+		System.out.println(ufprLearningCurveUtils.strMainStatisticsMatlab(avgResult));
+	}
+
+
+}