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

Example 6 with Attribute

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

the class ArffParserTest method testParseIris.

/**
     * Test of parse method, of class ArffParser.
     */
@Test
public void testParseIris() throws Exception {
    System.out.println("iris");
    try {
        ArffParser arffParser = new ArffParser();
        arffParser.setResponseIndex(4);
        AttributeDataset iris = arffParser.parse(smile.data.parser.IOUtils.getTestDataFile("weka/iris.arff"));
        double[][] x = iris.toArray(new double[iris.size()][]);
        int[] y = iris.toArray(new int[iris.size()]);
        assertEquals(Attribute.Type.NOMINAL, iris.response().getType());
        for (Attribute attribute : iris.attributes()) {
            assertEquals(Attribute.Type.NUMERIC, attribute.getType());
        }
        assertEquals(150, iris.size());
        assertEquals(4, iris.attributes().length);
        assertEquals("Iris-setosa", iris.response().toString(y[0]));
        assertEquals("Iris-setosa", iris.response().toString(y[1]));
        assertEquals("Iris-setosa", iris.response().toString(y[2]));
        assertEquals(5.1, x[0][0], 1E-7);
        assertEquals(3.5, x[0][1], 1E-7);
        assertEquals(1.4, x[0][2], 1E-7);
        assertEquals(0.2, x[0][3], 1E-7);
        assertEquals("Iris-virginica", iris.response().toString(y[149]));
        assertEquals(5.9, x[149][0], 1E-7);
        assertEquals(3.0, x[149][1], 1E-7);
        assertEquals(5.1, x[149][2], 1E-7);
        assertEquals(1.8, x[149][3], 1E-7);
    } catch (Exception ex) {
        System.err.println(ex);
    }
}

Example 7 with Attribute

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

the class ArffParser method getAttributes.

/**
     * Returns the attribute set of given stream.
     */
public static Attribute[] getAttributes(InputStream stream) throws IOException, ParseException {
    Reader r = new BufferedReader(new InputStreamReader(stream));
    StreamTokenizer tokenizer = new StreamTokenizer(r);
    ArffParser parser = new ArffParser();
    parser.initTokenizer(tokenizer);
    List<Attribute> attributes = new ArrayList<>();
    parser.readHeader(tokenizer, attributes);
    return attributes.toArray(new Attribute[attributes.size()]);
}

Example 8 with Attribute

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

the class ArffParser method parse.

/**
     * Parse a dataset from given stream.
     */
public AttributeDataset parse(InputStream stream) throws IOException, ParseException {
    try (Reader r = new BufferedReader(new InputStreamReader(stream))) {
        StreamTokenizer tokenizer = new StreamTokenizer(r);
        initTokenizer(tokenizer);
        List<Attribute> attributes = new ArrayList<>();
        String relationName = readHeader(tokenizer, attributes);
        if (attributes.isEmpty()) {
            throw new IOException("no header information available");
        }
        Attribute response = null;
        Attribute[] attr = new Attribute[attributes.size()];
        attributes.toArray(attr);
        for (int i = 0; i < attributes.size(); i++) {
            if (responseIndex == i) {
                response = attributes.remove(i);
                break;
            }
        }
        AttributeDataset data = new AttributeDataset(relationName, attributes.toArray(new Attribute[attributes.size()]), response);
        while (true) {
            // Check if end of file reached.
            getFirstToken(tokenizer);
            if (tokenizer.ttype == StreamTokenizer.TT_EOF) {
                break;
            }
            // Parse instance
            if (tokenizer.ttype == '{') {
                data.add(getSparseInstance(tokenizer, attr));
            } else {
                data.add(getInstance(tokenizer, attr));
            }
        }
        for (Attribute attribute : attributes) {
            if (attribute instanceof NominalAttribute) {
                NominalAttribute a = (NominalAttribute) attribute;
                a.setOpen(false);
            }
            if (attribute instanceof StringAttribute) {
                StringAttribute a = (StringAttribute) attribute;
                a.setOpen(false);
            }
        }
        return data;
    }
}

Example 9 with Attribute

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

the class LibsvmParser method parse.

/**
     * Parse a libsvm sparse dataset from an input stream.
     * @param name the name of dataset.
     * @param stream the input stream of data.
     * @throws java.io.IOException
     */
public SparseDataset parse(String name, InputStream stream) throws IOException, ParseException {
    BufferedReader reader = new BufferedReader(new InputStreamReader(stream));
    try {
        String line = reader.readLine();
        if (line == null) {
            throw new IOException("Empty data source.");
        }
        String[] tokens = line.trim().split("\\s+");
        boolean classification = true;
        Attribute response = null;
        try {
            Integer.valueOf(tokens[0]);
            response = new NominalAttribute("class");
        } catch (NumberFormatException e) {
            try {
                Double.valueOf(tokens[0]);
                response = new NominalAttribute("response");
                classification = false;
            } catch (NumberFormatException ex) {
                logger.error("Failed to parse {}", tokens[0], ex);
                throw new NumberFormatException("Unrecognized response variable value: " + tokens[0]);
            }
        }
        SparseDataset sparse = new SparseDataset(name, response);
        for (int i = 0; line != null; i++) {
            tokens = line.trim().split("\\s+");
            if (classification) {
                int y = Integer.parseInt(tokens[0]);
                sparse.set(i, y);
            } else {
                double y = Double.parseDouble(tokens[0]);
                sparse.set(i, y);
            }
            for (int k = 1; k < tokens.length; k++) {
                String[] pair = tokens[k].split(":");
                if (pair.length != 2) {
                    throw new NumberFormatException("Invalid data: " + tokens[k]);
                }
                int j = Integer.parseInt(pair[0]) - 1;
                double x = Double.parseDouble(pair[1]);
                sparse.set(i, j, x);
            }
            line = reader.readLine();
        }
        if (classification) {
            int n = sparse.size();
            int[] y = sparse.toArray(new int[n]);
            int[] label = Math.unique(y);
            Arrays.sort(label);
            for (int c : label) {
                response.valueOf(String.valueOf(c));
            }
            for (int i = 0; i < n; i++) {
                sparse.get(i).y = Arrays.binarySearch(label, y[i]);
            }
        }
        return sparse;
    } finally {
        reader.close();
    }
}

Example 10 with Attribute

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

the class SammonMappingDemo method learn.

/**
     * Execute the MDS algorithm and return a swing JComponent representing
     * the clusters.
     */
public JComponent learn() {
    JPanel pane = new JPanel(new GridLayout(1, 2));
    double[][] data = dataset[datasetIndex].toArray(new double[dataset[datasetIndex].size()][]);
    String[] labels = dataset[datasetIndex].toArray(new String[dataset[datasetIndex].size()]);
    if (labels[0] == null) {
        Attribute[] attr = dataset[datasetIndex].attributes();
        labels = new String[attr.length];
        for (int i = 0; i < labels.length; i++) {
            labels[i] = attr[i].getName();
        }
    }
    long clock = System.currentTimeMillis();
    SammonMapping sammon = new SammonMapping(data, 2);
    System.out.format("Learn Sammon's Mapping (k=2) from %d samples in %dms\n", data.length, System.currentTimeMillis() - clock);
    PlotCanvas plot = ScatterPlot.plot(sammon.getCoordinates(), labels);
    plot.setTitle("Sammon's Mapping (k = 2)");
    pane.add(plot);
    clock = System.currentTimeMillis();
    sammon = new SammonMapping(data, 3);
    System.out.format("Learn Sammon's Mapping (k=3) from %d samples in %dms\n", data.length, System.currentTimeMillis() - clock);
    plot = ScatterPlot.plot(sammon.getCoordinates(), labels);
    plot.setTitle("Sammon's Mapping (k = 3)");
    pane.add(plot);
    return pane;
}