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

Example 11 with AttributeDataset

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

the class GrowingNeuralGasTest method testUSPS.

/**
     * Test of learn method, of class GrowingNeuralGas.
     */
@Test
public void testUSPS() {
    System.out.println("USPS");
    DelimitedTextParser parser = new DelimitedTextParser();
    parser.setResponseIndex(new NominalAttribute("class"), 0);
    try {
        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()]);
        GrowingNeuralGas gng = new GrowingNeuralGas(x[0].length);
        for (int i = 0; i < 10; i++) {
            int[] index = Math.permutate(x.length);
            for (int j = 0; j < x.length; j++) {
                gng.update(x[index[j]]);
            }
        }
        gng.partition(10);
        AdjustedRandIndex ari = new AdjustedRandIndex();
        RandIndex rand = new RandIndex();
        int[] p = new int[x.length];
        for (int i = 0; i < x.length; i++) {
            p[i] = gng.predict(x[i]);
        }
        double r = rand.measure(y, p);
        double r2 = ari.measure(y, p);
        System.out.format("Training rand index = %.2f%%\tadjusted rand index = %.2f%%%n", 100.0 * r, 100.0 * r2);
        assertTrue(r > 0.85);
        assertTrue(r2 > 0.40);
        p = new int[testx.length];
        for (int i = 0; i < testx.length; i++) {
            p[i] = gng.predict(testx[i]);
        }
        r = rand.measure(testy, p);
        r2 = ari.measure(testy, p);
        System.out.format("Testing rand index = %.2f%%\tadjusted rand index = %.2f%%%n", 100.0 * r, 100.0 * r2);
        assertTrue(r > 0.85);
        assertTrue(r2 > 0.40);
    } catch (Exception ex) {
        System.err.println(ex);
    }
}

Example 12 with AttributeDataset

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

the class NeuralMapTest method testUSPS.

/**
     * Test of learn method, of class NeuralMap.
     */
@Test
public void testUSPS() {
    System.out.println("USPS");
    DelimitedTextParser parser = new DelimitedTextParser();
    parser.setResponseIndex(new NominalAttribute("class"), 0);
    try {
        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()]);
        NeuralMap cortex = new NeuralMap(x[0].length, 8.0, 0.05, 0.0006, 5, 3);
        for (int i = 0; i < 5; i++) {
            for (double[] xi : x) {
                cortex.update(xi);
            }
        }
        cortex.purge(16);
        cortex.partition(10);
        AdjustedRandIndex ari = new AdjustedRandIndex();
        RandIndex rand = new RandIndex();
        int[] p = new int[x.length];
        for (int i = 0; i < x.length; i++) {
            p[i] = cortex.predict(x[i]);
        }
        double r = rand.measure(y, p);
        double r2 = ari.measure(y, p);
        System.out.format("Training rand index = %.2f%%\tadjusted rand index = %.2f%%%n", 100.0 * r, 100.0 * r2);
        //assertTrue(r > 0.65);
        //assertTrue(r2 > 0.18);
        p = new int[testx.length];
        for (int i = 0; i < testx.length; i++) {
            p[i] = cortex.predict(testx[i]);
        }
        r = rand.measure(testy, p);
        r2 = ari.measure(testy, p);
        System.out.format("Testing rand index = %.2f%%\tadjusted rand index = %.2f%%%n", 100.0 * r, 100.0 * r2);
    //assertTrue(r > 0.65);
    //assertTrue(r2 > 0.18);
    } catch (Exception ex) {
        System.err.println(ex);
    }
}

Example 13 with AttributeDataset

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

the class SOMTest method testUSPS.

/**
     * Test of learn method, of class SOM.
     */
@Test
public void testUSPS() {
    System.out.println("USPS");
    DelimitedTextParser parser = new DelimitedTextParser();
    parser.setResponseIndex(new NominalAttribute("class"), 0);
    try {
        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()]);
        SOM som = new SOM(x, 10, 10);
        int[] label = som.partition(10);
        AdjustedRandIndex ari = new AdjustedRandIndex();
        RandIndex rand = new RandIndex();
        double r = rand.measure(y, label);
        double r2 = ari.measure(y, label);
        System.out.format("Training rand index = %.2f%%\tadjusted rand index = %.2f%%%n", 100.0 * r, 100.0 * r2);
        assertTrue(r > 0.88);
        assertTrue(r2 > 0.45);
        int[] p = new int[testx.length];
        for (int i = 0; i < testx.length; i++) {
            p[i] = som.predict(testx[i]);
        }
        r = rand.measure(testy, p);
        r2 = ari.measure(testy, p);
        System.out.format("Testing rand index = %.2f%%\tadjusted rand index = %.2f%%%n", 100.0 * r, 100.0 * r2);
        assertTrue(r > 0.88);
        assertTrue(r2 > 0.45);
    } catch (Exception ex) {
        System.err.println(ex);
    }
}

Example 14 with AttributeDataset

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

the class GCTParser method parse.

/**
     * Parse a GCT dataset from an input stream.
     * @param name the name of dataset.
     * @param stream the input stream of data.
     * @throws java.io.IOException
     */
public AttributeDataset parse(String name, InputStream stream) throws IOException, ParseException {
    BufferedReader reader = new BufferedReader(new InputStreamReader(stream));
    String line = reader.readLine();
    if (line == null) {
        throw new IOException("Empty data source.");
    }
    if (!line.equals("#1.2")) {
        throw new IOException("Invalid version.");
    }
    line = reader.readLine();
    if (line == null) {
        throw new IOException("Premature end of file.");
    }
    String[] tokens = line.split("\t", -1);
    if (tokens.length != 2) {
        throw new IOException("Invalid data size inforamation.");
    }
    int n = Integer.parseInt(tokens[0]);
    int p = Integer.parseInt(tokens[1]);
    if (n <= 0 || p <= 0) {
        throw new IOException(String.format("Invalid data size %d x %d.", n, p));
    }
    Attribute[] attributes = new Attribute[p];
    line = reader.readLine();
    if (line == null) {
        throw new IOException("Premature end of file.");
    }
    tokens = line.split("\t", -1);
    if (tokens.length != p + 2) {
        throw new IOException("Invalid title header.");
    }
    for (int i = 0; i < p; i++) {
        attributes[i] = new NumericAttribute(tokens[i + 2]);
    }
    AttributeDataset data = new AttributeDataset(name, attributes);
    for (int i = 0; i < n; i++) {
        line = reader.readLine();
        if (line == null) {
            throw new IOException("Premature end of file.");
        }
        tokens = line.split("\t", -1);
        if (tokens.length != p + 2) {
            throw new IOException(String.format("Invalid number of elements of line %d: %d", i + 4, tokens.length));
        }
        double[] x = new double[p];
        for (int j = 0; j < p; j++) {
            if (tokens[j + 2].isEmpty()) {
                x[j] = Double.NaN;
            } else {
                x[j] = Double.valueOf(tokens[j + 2]);
            }
        }
        Datum<double[]> datum = new Datum<>(x);
        datum.name = tokens[0];
        datum.description = tokens[1];
        data.add(datum);
    }
    reader.close();
    return data;
}

Example 15 with AttributeDataset

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

the class PCLParser method parse.

/**
     * Parse a PCL dataset from an input stream.
     * @param name the name of dataset.
     * @param stream the input stream of data.
     * @throws java.io.IOException
     */
public AttributeDataset parse(String name, InputStream stream) throws IOException, ParseException {
    BufferedReader reader = new BufferedReader(new InputStreamReader(stream));
    String line = reader.readLine();
    if (line == null) {
        throw new IOException("Empty data source.");
    }
    String[] tokens = line.split("\t", -1);
    int p = tokens.length - 3;
    line = reader.readLine();
    if (line == null) {
        throw new IOException("Premature end of file.");
    }
    String[] weight = line.split("\t", -1);
    if (weight.length != tokens.length) {
        throw new IOException("Invalid sample weight header.");
    }
    Attribute[] attributes = new Attribute[p];
    for (int i = 0; i < p; i++) {
        attributes[i] = new NumericAttribute(tokens[i + 3], null, Double.valueOf(weight[i + 3]));
    }
    AttributeDataset data = new AttributeDataset(name, attributes);
    for (int i = 3; (line = reader.readLine()) != null; i++) {
        tokens = line.split("\t", -1);
        if (tokens.length != weight.length) {
            throw new IOException(String.format("Invalid number of elements of line %d: %d", i, tokens.length));
        }
        double[] x = new double[p];
        for (int j = 0; j < p; j++) {
            if (tokens[j + 3].isEmpty()) {
                x[j] = Double.NaN;
            } else {
                x[j] = Double.valueOf(tokens[j + 3]);
            }
        }
        Datum<double[]> datum = new Datum<>(x);
        datum.name = tokens[0];
        datum.description = tokens[1];
        datum.weight = Double.valueOf(tokens[2]);
        data.add(datum);
    }
    reader.close();
    return data;
}