Java Kernel类代码示例

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
/**
 * This method will be executed before each test method.
 */
@Before
public void initializeExamples() {
	String reprA = "fakeclass |BDV:"+DENSE_NAME+"| 1.0,0.0,1.0 |EV| |BV:"+SPARSE_NAME+"| one:1.0 three:1.0 |EV|";
	String reprB = "fakeclass |BDV:"+DENSE_NAME+"| 0.0,1.0,1.0 |EV| |BV:"+SPARSE_NAME+"| two:1.0 three:1.0 |EV|";

	try {
		a = ExampleFactory.parseExample(reprA);
		b = ExampleFactory.parseExample(reprB);
	} catch (InstantiationException e) {
		e.printStackTrace();
		Assert.fail();
	}
	
	Kernel kernelOnDense = new PolynomialKernel(2, new LinearKernel(DENSE_NAME));
	Kernel kernelOnSparse = new NormalizationKernel(new LinearKernel(SPARSE_NAME));
	kernel = new LinearKernelCombination();
	kernel.addKernel(1, kernelOnDense);
	kernel.addKernel(2, kernelOnSparse);
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
/**
 * This method will be executed before each test method.
 */
@Before
public void initializeExamples() {
	String textualA = "class1 class2 |<| lab1 |BDV:" + DENSE_NAME + "| 0.5 1 |EDV| |,| |BDV:" + DENSE_NAME + "| -3 1 |EDV| |>| |BV:" + SPARSE_NAME + "| ptkSim:0.8 |EV|";
	String textualB = "class1 class2 |<| lab1 |BDV:" + DENSE_NAME + "| -0.5 1.3 |EDV| |,| |BDV:" + DENSE_NAME + "| 0 1 |EDV| |>| |BV:" + SPARSE_NAME + "| ptkSim:0.3 |EV|";
	

	try {
		pairA = ExampleFactory.parseExample(textualA);
		pairB = ExampleFactory.parseExample(textualB);
	} catch (InstantiationException e) {
		e.printStackTrace();
		Assert.fail();
	}


	Kernel kernelOnDense = new LinearKernel(DENSE_NAME);
	PreferenceKernel preference = new PreferenceKernel(kernelOnDense);
	Kernel kernelOnSparse = new LinearKernel(SPARSE_NAME);
	kernel = new LinearKernelCombination();
	kernel.addKernel(1, preference);
	kernel.addKernel(2, kernelOnSparse);
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
private static float tune(SimpleDataset allTrainingSet, Kernel kernel,
		float split, float[] cs) throws NoSuchPerformanceMeasureException,
		IOException {
	float bestC = 0.0f;
	float bestF1 = -Float.MAX_VALUE;

	// Split data according to a fix split
	Dataset[] split2 = allTrainingSet
			.splitClassDistributionInvariant(split);
	SimpleDataset trainingSet = (SimpleDataset) split2[0];
	SimpleDataset testSet = (SimpleDataset) split2[1];
	// tune parameter C
	for (float c : cs) {
		float f1 = test(trainingSet, kernel, c, testSet, false);
		System.out.println("C:" + c + "\t" + f1);
		if (f1 > bestF1) {
			bestF1 = f1;
			bestC = c;
		}
	}

	return bestC;
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
@BeforeClass
public static void learnModel() {
	trainingSet = new SimpleDataset();
	testSet = new SimpleDataset();
	try {
		trainingSet.populate("src/test/resources/svmTest/binary/binary_train.klp");
		// Read a dataset into a test variable
		testSet.populate("src/test/resources/svmTest/binary/binary_test.klp");
	} catch (Exception e) {
		e.printStackTrace();
		Assert.assertTrue(false);
	}

	// define the positive class
	StringLabel positiveClass = new StringLabel("+1");

	// define the kernel
	Kernel kernel = new LinearKernel("0");

	// add a cache
	kernel.setKernelCache(new FixIndexKernelCache(trainingSet
			.getNumberOfExamples()));

	// define the learning algorithm
	BinaryNuSvmClassification learner = new BinaryNuSvmClassification(kernel,
			positiveClass, 0.5f);

	// learn and get the prediction function
	learner.learn(trainingSet);
	f = learner.getPredictionFunction();
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
@BeforeClass
public static void learnModel() {
	trainingSet = new SimpleDataset();
	testSet = new SimpleDataset();
	try {
		trainingSet.populate("src/test/resources/svmTest/binary/binary_train.klp");
		// Read a dataset into a test variable
		testSet.populate("src/test/resources/svmTest/binary/binary_test.klp");
	} catch (Exception e) {
		e.printStackTrace();
		Assert.assertTrue(false);
	}

	// define the positive class
	StringLabel positiveClass = new StringLabel("+1");

	// define the kernel
	Kernel kernel = new LinearKernel("0");

	// add a cache
	kernel.setKernelCache(new FixIndexKernelCache(trainingSet
			.getNumberOfExamples()));

	// define the learning algorithm
	BinaryCSvmClassification learner = new BinaryCSvmClassification(kernel,
			positiveClass, 1, 1);

	// learn and get the prediction function
	learner.learn(trainingSet);
	f = learner.getPredictionFunction();
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
@BeforeClass
public static void learnModel() {
	trainingSet = new SimpleDataset();
	testSet = new SimpleDataset();
	try {
		trainingSet
				.populate("src/test/resources/svmTest/regression/mg_scale.klp");
		// Read a dataset into a test variable
		testSet.populate("src/test/resources/svmTest/regression/mg_scale.klp");
	} catch (Exception e) {
		e.printStackTrace();
		Assert.assertTrue(false);
	}

	// define the regression label
	Label label = new StringLabel("r");

	// define the kernel
	Kernel kernel = new LinearKernel("0");

	// add a cache
	kernel.setKernelCache(new FixIndexKernelCache(trainingSet
			.getNumberOfExamples()));

	// define the learning algorithm
	EpsilonSvmRegression learner = new EpsilonSvmRegression(kernel, label,
			1, 0.1f);

	// learn and get the prediction function
	learner.learn(trainingSet);
	p = learner.getPredictionFunction();
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
public Kernel getKernel() {
	return kernel;
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
public void setKernel(Kernel kernel) {
	this.kernel = kernel;
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
public static void main(String[] args) throws Exception {
	float split = 0.8f;
	String train_file = "src/main/resources/tweetSentiment2013/train.klp.gz";
	String test_file = "src/main/resources/tweetSentiment2013/test.klp.gz";
	int kernelmode = 1;
	float polyD = 0;
	float gamma = 0;

	float[] Cs = new float[] { 0.1f, 0.5f, 1f };

	// Read a dataset into a test variable
	SimpleDataset trainingSet = new SimpleDataset();
	trainingSet.populate(train_file);
	// Read a dataset into a test variable
	SimpleDataset testSet = new SimpleDataset();
	testSet.populate(test_file);
	// set the cache size
	int cacheSize = trainingSet.getNumberOfExamples()
			+ testSet.getNumberOfExamples();
	// Initialize a kernel
	Kernel kernel = null;
	switch (kernelmode) {
	case 1:
		kernel = getBowKernel(cacheSize);
		break;
	case 2:
		kernel = getPolyBow(cacheSize, polyD);
		break;
	case 3:
		kernel = getWordspaceKernel(cacheSize);
		break;
	case 4:
		kernel = getRbfWordspaceKernel(cacheSize, gamma);
		break;
	case 5:
		kernel = getBowWordSpaceKernel(cacheSize);
		break;
	case 6:
		kernel = getPolyBowRbfWordspaceKernel(cacheSize, polyD, gamma);
		break;
	default:
		kernel = getBowKernel(cacheSize);
		break;
	}

	// Find optimal C
	float c = tune(trainingSet, kernel, split, Cs);
	System.out.println("start testing with C=" + c);
	// test
	float f1 = test(trainingSet, kernel, c, testSet, true);
	System.out.println("Mean F1 on test set=" + f1);
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
public static void main(String[] args) throws Exception {
	// The epsilon in loss function of the regressor
	float pReg = 0.1f;
	// The regularization parameter of the regressor
	float c = 2f;
	// The gamma parameter in the RBF kernel
	float gamma = 1f;

	// The label indicating the value considered by the regressor
	Label label = new StringLabel("r");

	// Load the dataset
	SimpleDataset dataset = new SimpleDataset();
	dataset.populate("src/main/resources/sv_regression_test/mg_scale.klp");
	// Split the dataset in train and test datasets
	dataset.shuffleExamples(new Random(0));
	SimpleDataset[] split = dataset.split(0.7f);
	SimpleDataset trainDataset = split[0];
	SimpleDataset testDataset = split[1];

	// Kernel for the first representation (0-index)
	Kernel linear = new LinearKernel("0");
	// Applying the RBF kernel
	Kernel rbf = new RbfKernel(gamma, linear);
	// Applying a cache
	FixIndexKernelCache kernelCache = new FixIndexKernelCache(
			trainDataset.getNumberOfExamples());
	rbf.setKernelCache(kernelCache);

	// instantiate the regressor
	EpsilonSvmRegression regression = new EpsilonSvmRegression(rbf, label,
			c, pReg);

	// learn
	regression.learn(trainDataset);
	// get the prediction function
	RegressionFunction regressor = regression.getPredictionFunction();

	// initializing the performance evaluator
	RegressorEvaluator evaluator = new RegressorEvaluator(
			trainDataset.getRegressionProperties());

	// For each example from the test set
	for (Example e : testDataset.getExamples()) {
		// Predict the value
		Prediction prediction = regressor.predict(e);
		// Print the original and the predicted values
		System.out.println("real value: " + e.getRegressionValue(label)
				+ "\t-\tpredicted value: " + prediction.getScore(label));
		// Update the evaluator
		evaluator.addCount(e, prediction);
	}

	// Get the Mean Squared Error for the targeted label
	float measSquareError = evaluator.getMeanSquaredError(label);

	System.out.println("\nMean Squared Error:\t" + measSquareError);
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
public static void main(String[] args) {
	try {
		// Read a dataset into a trainingSet variable
		SimpleDataset trainingSet = new SimpleDataset();
		trainingSet.populate("src/main/resources/multiplerepresentation/train.klp");
		// Read a dataset into a test variable
		SimpleDataset testSet = new SimpleDataset();
		testSet.populate("src/main/resources/multiplerepresentation/test.klp");

		List<Label> classes = trainingSet.getClassificationLabels();

		
		for (int i=0; i<classes.size(); ++i) {
			Label l = classes.get(i);
			System.out.println("Class: " + l.toString());
			System.out.println(trainingSet.getNumberOfPositiveExamples(l));
			System.out.println(testSet.getNumberOfPositiveExamples(l));
		}
		
		// instantiate a passive aggressive algorithm
		KernelizedPassiveAggressiveClassification kPA = new KernelizedPassiveAggressiveClassification();
		// set an aggressiveness parameter
		kPA.setC(2f);

		// Kernel for the first representation (0-index)
		Kernel linear = new LinearKernel("0");
		// Normalize the linear kernel
		NormalizationKernel normalizedKernel = new NormalizationKernel(
				linear);
		// Apply a 2-degree Polynomial kernel on the score (normalized) computed by
		// the linear kernel
		Kernel polyKernel = new PolynomialKernel(2f, normalizedKernel);

		// Kernel for the second representation (1-index)
		Kernel linear1 = new LinearKernel("1");
		// Normalize the linear kernel
		NormalizationKernel normalizedKernel1 = new NormalizationKernel(
				linear1);
		// Apply a RBF kernel on the score (normalized) computed by
		// the linear kernel
		Kernel rbfKernel = new RbfKernel(2f, normalizedKernel1);
		// tell the algorithm that the kernel we want to use in learning is
		// the polynomial kernel

		LinearKernelCombination linearCombination = new LinearKernelCombination();
		linearCombination.addKernel(1f, polyKernel);
		linearCombination.addKernel(1f, rbfKernel);
		// normalize the weights such that their sum is 1
		linearCombination.normalizeWeights();
		
		// set the kernel for the PA algorithm
		kPA.setKernel(linearCombination);
		
		// Instantiate a OneVsAll learning algorithm
		// It is a so called meta learner, it receives in input a binary learning algorithm
		OneVsAllLearning metaOneVsAllLearner = new OneVsAllLearning();
		metaOneVsAllLearner.setBaseAlgorithm(kPA);
		metaOneVsAllLearner.setLabels(classes);

		long startLearningTime = System.currentTimeMillis();
		// learn and get the prediction function
		metaOneVsAllLearner.learn(trainingSet);
		OneVsAllClassifier f = metaOneVsAllLearner.getPredictionFunction();
		long endLearningTime = System.currentTimeMillis();

		// classify examples and compute some statistics
		MulticlassClassificationEvaluator ev = new MulticlassClassificationEvaluator(classes);
		for (Example e : testSet.getExamples()) {
			OneVsAllClassificationOutput prediction = f.predict(e);
			ev.addCount(e, prediction);
		}

		System.out
				.println("Accuracy: "
						+ ev.getAccuracy());
		System.out.println("Learning time without cache: " + (endLearningTime-startLearningTime) + " ms");
	} catch (Exception e1) {
		e1.printStackTrace();
	}
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
public static void main(String[] args) {
	try {
		// Read a dataset into a trainingSet variable
		SimpleDataset trainingSet = new SimpleDataset();
		trainingSet.populate("src/main/resources/hellolearning/train.klp");
		// Read a dataset into a test variable
		SimpleDataset testSet = new SimpleDataset();
		testSet.populate("src/main/resources/hellolearning/test.klp");

		// define the positive class
		StringLabel positiveClass = new StringLabel("+1");

		// print some statistics
		System.out.println("Training set statistics");
		System.out.print("Examples number ");
		System.out.println(trainingSet.getNumberOfExamples());
		System.out.print("Positive examples ");
		System.out.println(trainingSet
				.getNumberOfPositiveExamples(positiveClass));
		System.out.print("Negative examples ");
		System.out.println(trainingSet
				.getNumberOfNegativeExamples(positiveClass));

		System.out.println("Test set statistics");
		System.out.print("Examples number ");
		System.out.println(testSet.getNumberOfExamples());
		System.out.print("Positive examples ");
		System.out.println(testSet
				.getNumberOfPositiveExamples(positiveClass));
		System.out.print("Negative examples ");
		System.out.println(testSet
				.getNumberOfNegativeExamples(positiveClass));

		// instantiate a passive aggressive algorithm
		KernelizedPassiveAggressiveClassification kPA = new KernelizedPassiveAggressiveClassification();
		// indicate to the learner what is the positive class
		kPA.setLabel(positiveClass);
		// set an aggressiveness parameter
		kPA.setC(0.01f);

		// use the first (and only here) representation
		Kernel linear = new LinearKernel("0");
		// Normalize the linear kernel
		NormalizationKernel normalizedKernel = new NormalizationKernel(
				linear);
		// Apply a Polynomial kernel on the score (normalized) computed by
		// the linear kernel
		Kernel polyKernel = new PolynomialKernel(2f, normalizedKernel);
		// tell the algorithm that the kernel we want to use in learning is
		// the polynomial kernel
		kPA.setKernel(polyKernel);

		// learn and get the prediction function
		kPA.learn(trainingSet);
		Classifier f = kPA.getPredictionFunction();
		// classify examples and compute some statistics
		BinaryClassificationEvaluator ev = new BinaryClassificationEvaluator(positiveClass);
		for (Example e : testSet.getExamples()) {
			ClassificationOutput p = f.predict(testSet.getNextExample());
			ev.addCount(e, p);
		}

		System.out
				.println("Accuracy: " +
						ev.getAccuracy());
	} catch (Exception e1) {
		e1.printStackTrace();
	}
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
public static void main(String[] args) {
	try {
		// Read a dataset into a trainingSet variable
		SimpleDataset trainingSet = new SimpleDataset();
		trainingSet
				.populate("src/main/resources/sequenceKernelExample/sequenceTrain.txt");

		SimpleDataset testSet = new SimpleDataset();
		testSet.populate("src/main/resources/sequenceKernelExample/sequenceTest.txt");

		// print some statistics
		System.out.println("Training set statistics");
		System.out.print("Examples number ");
		System.out.println(trainingSet.getNumberOfExamples());

		List<Label> classes = trainingSet.getClassificationLabels();

		for (Label l : classes) {
			System.out.println("Training Label " + l.toString() + " "
					+ trainingSet.getNumberOfPositiveExamples(l));
			System.out.println("Training Label " + l.toString() + " "
					+ trainingSet.getNumberOfNegativeExamples(l));

			System.out.println("Test Label " + l.toString() + " "
					+ testSet.getNumberOfPositiveExamples(l));
			System.out.println("Test Label " + l.toString() + " "
					+ testSet.getNumberOfNegativeExamples(l));
		}

		// Kernel for the first representation (0-index)
		Kernel kernel = new SequenceKernel("SEQUENCE", 2, 1);
		// Normalize the linear kernel
		NormalizationKernel normalizedKernel = new NormalizationKernel(
				kernel);
		kernel.setSquaredNormCache(new FixIndexSquaredNormCache(trainingSet.getNumberOfExamples()));
		kernel.setKernelCache(new FixIndexKernelCache(trainingSet.getNumberOfExamples()));
		// instantiate an svmsolver
		BinaryCSvmClassification svmSolver = new BinaryCSvmClassification();
		svmSolver.setKernel(normalizedKernel);
		svmSolver.setCp(1);
		svmSolver.setCn(1);

		OneVsAllLearning ovaLearner = new OneVsAllLearning();
		ovaLearner.setBaseAlgorithm(svmSolver);
		ovaLearner.setLabels(classes);

		// learn and get the prediction function
		ovaLearner.learn(trainingSet);
		Classifier f = ovaLearner.getPredictionFunction();

		// classify examples and compute some statistics
		MulticlassClassificationEvaluator ev = new MulticlassClassificationEvaluator(
				trainingSet.getClassificationLabels());

		for (Example e : testSet.getExamples()) {
			ClassificationOutput p = f.predict(testSet.getNextExample());
			ev.addCount(e, p);
		}

		System.out.println("Accuracy: "
				+ ev.getPerformanceMeasure("accuracy"));
	} catch (Exception e1) {
		e1.printStackTrace();
	}
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
public static void main(String[] args) {
	try {
		// Read a dataset into a trainingSet variable
		SimpleDataset trainingSet = new SimpleDataset();
		trainingSet
				.populate("src/main/resources/iris_dataset/iris_train.klp");

		SimpleDataset testSet = new SimpleDataset();
		testSet.populate("src/main/resources/iris_dataset/iris_test.klp");

		// print some statistics
		System.out.println("Training set statistics");
		System.out.print("Examples number ");
		System.out.println(trainingSet.getNumberOfExamples());

		List<Label> classes = trainingSet.getClassificationLabels();

		for (Label l : classes) {
			System.out.println("Training Label " + l.toString() + " "
					+ trainingSet.getNumberOfPositiveExamples(l));
			System.out.println("Training Label " + l.toString() + " "
					+ trainingSet.getNumberOfNegativeExamples(l));

			System.out.println("Test Label " + l.toString() + " "
					+ testSet.getNumberOfPositiveExamples(l));
			System.out.println("Test Label " + l.toString() + " "
					+ testSet.getNumberOfNegativeExamples(l));
		}

		// Kernel for the first representation (0-index)
		Kernel linear = new LinearKernel("0");
		// Normalize the linear kernel
		NormalizationKernel normalizedKernel = new NormalizationKernel(
				linear);
		// instantiate an svmsolver
		BinaryCSvmClassification svmSolver = new BinaryCSvmClassification();
		svmSolver.setKernel(normalizedKernel);
		svmSolver.setCp(1);
		svmSolver.setCn(1);

		OneVsAllLearning ovaLearner = new OneVsAllLearning();
		ovaLearner.setBaseAlgorithm(svmSolver);
		ovaLearner.setLabels(classes);

		// learn and get the prediction function
		ovaLearner.learn(trainingSet);
		Classifier f = ovaLearner.getPredictionFunction();

		// classify examples and compute some statistics
		MulticlassClassificationEvaluator ev = new MulticlassClassificationEvaluator(
				trainingSet.getClassificationLabels());

		for (Example e : testSet.getExamples()) {
			ClassificationOutput p = f.predict(testSet.getNextExample());
			ev.addCount(e, p);
		}

		List<Label> twoLabels = new ArrayList<Label>();
		twoLabels.add(new StringLabel("iris-setosa"));
		twoLabels.add(new StringLabel("iris-virginica"));

		Object[] as = new Object[1];
		as[0] = twoLabels;

		System.out.println("Mean F1: "
				+ ev.getPerformanceMeasure("MeanF1"));
		System.out.println("Mean F1 For iris-setosa/iris-virginica: "
				+ ev.getPerformanceMeasure("MeanF1For", as));
		System.out.println("F1: "
				+ ev.getPerformanceMeasure("OverallF1"));
	} catch (Exception e1) {
		e1.printStackTrace();
	}
}
 

import it.uniroma2.sag.kelp.kernel.Kernel; //导入依赖的package包/类
/**
 * @param kernel
 *            The kernel function
 * @param k
 *            The number of expected clusters
 * @param maxIterations
 *            The maximum number of iterations
 */
public KernelBasedKMeansEngine(Kernel kernel, int k, int maxIterations) {
	this();
	this.kernel = kernel;
	this.k = k;
	this.maxIterations = maxIterations;
}