Examples with AttributeDataset smile.data.AttributeDataset used on opensource projects

Example 61 with AttributeDataset

use of smile.data.AttributeDataset in project smile by haifengl.

the class Nominal2SparseBinaryTest method testF.

/**
     * Test of f method, of class Nominal2SparseBinary.
     */
@Test
public void testF() {
    System.out.println("f");
    int[][] result = { { 0, 3, 6, 9 }, { 0, 3, 6, 8 }, { 1, 3, 6, 9 }, { 2, 4, 6, 9 }, { 2, 5, 7, 9 }, { 2, 5, 7, 8 }, { 1, 5, 7, 8 }, { 0, 4, 6, 9 }, { 0, 5, 7, 9 }, { 2, 4, 7, 9 }, { 0, 4, 7, 8 }, { 1, 4, 6, 8 }, { 1, 3, 7, 9 }, { 2, 4, 6, 8 } };
    ArffParser arffParser = new ArffParser();
    arffParser.setResponseIndex(4);
    try {
        AttributeDataset weather = arffParser.parse(smile.data.parser.IOUtils.getTestDataFile("weka/weather.nominal.arff"));
        double[][] x = weather.toArray(new double[weather.size()][]);
        Nominal2SparseBinary n2sb = new Nominal2SparseBinary(weather.attributes());
        for (int i = 0; i < x.length; i++) {
            int[] y = n2sb.f(x[i]);
            assertEquals(result[i].length, y.length);
            for (int j = 0; j < y.length; j++) {
                assertEquals(result[i][j], y[j]);
            }
        }
    } catch (Exception ex) {
        System.err.println(ex);
    }
}

Example 62 with AttributeDataset

use of smile.data.AttributeDataset in project smile by haifengl.

the class NumericAttributeFeatureTest method testNORMALIZATION.

/**
     * Test of f method, of class NumericAttributeFeature.
     */
@Test
public void testNORMALIZATION() {
    System.out.println("NORMALIZATION");
    DelimitedTextParser parser = new DelimitedTextParser();
    parser.setResponseIndex(new NominalAttribute("class"), 0);
    try {
        AttributeDataset data = parser.parse("USPS Train", smile.data.parser.IOUtils.getTestDataFile("usps/zip.train"));
        double[][] x = data.toArray(new double[data.size()][]);
        double[] min = Math.colMin(x);
        double[] max = Math.colMax(x);
        NumericAttributeFeature naf = new NumericAttributeFeature(data.attributes(), NumericAttributeFeature.Scaling.NORMALIZATION, x);
        Attribute[] attributes = naf.attributes();
        assertEquals(256, attributes.length);
        for (int i = 0; i < x.length; i++) {
            double[] y = new double[attributes.length];
            for (int j = 0; j < y.length; j++) {
                y[j] = naf.f(x[i], j);
                assertEquals((x[i][j] - min[j]) / (max[j] - min[j]), y[j], 1E-7);
            }
        }
    } catch (Exception ex) {
        System.err.println(ex);
    }
}

Example 63 with AttributeDataset

use of smile.data.AttributeDataset in project smile by haifengl.

the class NumericAttributeFeatureTest method testNORMALIZATIONWinsorization.

/**
     * Test of f method, of class NumericAttributeFeature.
     */
@Test
public void testNORMALIZATIONWinsorization() {
    System.out.println("NORMALIZATION  Winsorization");
    ArffParser parser = new ArffParser();
    try {
        AttributeDataset data = parser.parse(smile.data.parser.IOUtils.getTestDataFile("weka/regression/abalone.arff"));
        double[][] x = data.toArray(new double[data.size()][]);
        NumericAttributeFeature naf = new NumericAttributeFeature(data.attributes(), 0.05, 0.95, x);
        Attribute[] attributes = naf.attributes();
        assertEquals(data.attributes().length - 1, attributes.length);
        for (int i = 0; i < x.length; i++) {
            double[] y = new double[attributes.length];
            for (int j = 0; j < y.length; j++) {
                y[j] = naf.f(x[i], j);
                assertTrue(y[j] <= 1.0 && y[j] >= 0.0);
            }
        }
    } catch (Exception ex) {
        System.err.println(ex);
    }
}

Example 64 with AttributeDataset

use of smile.data.AttributeDataset in project smile by haifengl.

the class NumericAttributeFeatureTest method testLOGARITHM.

/**
     * Test of f method, of class NumericAttributeFeature.
     */
@Test
public void testLOGARITHM() {
    System.out.println("LOGARITHM");
    DelimitedTextParser parser = new DelimitedTextParser();
    parser.setResponseIndex(new NominalAttribute("class"), 0);
    try {
        AttributeDataset data = parser.parse("USPS Train", smile.data.parser.IOUtils.getTestDataFile("usps/zip.train"));
        double[][] x = data.toArray(new double[data.size()][]);
        for (int i = 0; i < x.length; i++) {
            for (int j = 0; j < x[i].length; j++) {
                x[i][j] += 2.0;
            }
        }
        NumericAttributeFeature naf = new NumericAttributeFeature(data.attributes(), NumericAttributeFeature.Scaling.LOGARITHM);
        Attribute[] attributes = naf.attributes();
        assertEquals(256, attributes.length);
        for (int i = 0; i < x.length; i++) {
            double[] y = new double[attributes.length];
            for (int j = 0; j < y.length; j++) {
                y[j] = naf.f(x[i], j);
                assertEquals(Math.log(x[i][j]), y[j], 1E-7);
            }
        }
    } catch (Exception ex) {
        System.err.println(ex);
    }
}

Example 65 with AttributeDataset

use of smile.data.AttributeDataset in project smile by haifengl.

the class SumSquaresRatioTest method testLearn.

/**
     * Test of learn method, of class SumSquaresRatio.
     */
@Test
public void testLearn() {
    System.out.println("USPS");
    try {
        DelimitedTextParser parser = new DelimitedTextParser();
        parser.setResponseIndex(new NominalAttribute("class"), 0);
        AttributeDataset train = parser.parse("USPS Train", smile.data.parser.IOUtils.getTestDataFile("usps/zip.train"));
        AttributeDataset test = parser.parse("USPS Test", smile.data.parser.IOUtils.getTestDataFile("usps/zip.test"));
        double[][] x = train.toArray(new double[train.size()][]);
        int[] y = train.toArray(new int[train.size()]);
        double[][] testx = test.toArray(new double[test.size()][]);
        int[] testy = test.toArray(new int[test.size()]);
        SumSquaresRatio ssr = new SumSquaresRatio();
        double[] score = ssr.rank(x, y);
        int[] index = QuickSort.sort(score);
        int p = 135;
        int n = x.length;
        double[][] xx = new double[n][p];
        for (int j = 0; j < p; j++) {
            for (int i = 0; i < n; i++) {
                xx[i][j] = x[i][index[255 - j]];
            }
        }
        int testn = testx.length;
        double[][] testxx = new double[testn][p];
        for (int j = 0; j < p; j++) {
            for (int i = 0; i < testn; i++) {
                testxx[i][j] = testx[i][index[255 - j]];
            }
        }
        LDA lda = new LDA(xx, y);
        int[] prediction = new int[testn];
        for (int i = 0; i < testn; i++) {
            prediction[i] = lda.predict(testxx[i]);
        }
        double accuracy = new Accuracy().measure(testy, prediction);
        System.out.format("SSR %.2f%%%n", 100 * accuracy);
    } catch (Exception ex) {
        System.err.println(ex);
    }
}