org.dkpro.tc.ml.weka.util.MultilabelResult Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of dkpro-tc-ml-weka Show documentation
There is a newer version: 1.1.0
/**
 * Copyright 2018
 * Ubiquitous Knowledge Processing (UKP) Lab
 * Technische Universität Darmstadt
 *
 * 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 org.dkpro.tc.ml.weka.util;

import java.io.Serializable;
import java.util.Arrays;

import org.dkpro.tc.api.exception.TextClassificationException;

public class MultilabelResult
    implements Serializable
{

    private static final long serialVersionUID = -5066149060024459615L;

    /**
     * Wrapper for classification result from multi-label classification.
     * 
     * @param goldstandard
     *            the gold standard as integer matrix; labels x instances
     * @param predictions
     *            the predictions as double matrix; labels x instances
     * @param bipartitionThreshold
     *            a threshold to create bipartition from ranking
     * @throws TextClassificationException
     *             an exception
     */
    public MultilabelResult(int[][] goldstandard, double[][] predictions,
            String bipartitionThreshold)
        throws TextClassificationException
    {
        this.actuals = Arrays.copyOf(goldstandard, goldstandard.length);
        this.predictions = Arrays.copyOf(predictions, predictions.length);
        try {
            this.bipartitionThreshold = Double.parseDouble(bipartitionThreshold);
        }
        catch (NumberFormatException e) {
            // TODO
            throw new TextClassificationException(
                    "Currenty, only one global bipartition threshold value is supported. Please set a double as threshold.");
        }
    }

    /**
     * Returns the predictions
     * 
     * @return predictions as LxN matrix
     */
    public double[][] getPredictions()
    {
        return predictions.clone();
    }

    /**
     * Returns the gold standard
     * 
     * @return gold standard as LxN matrix
     */
    public int[][] getGoldstandard()
    {
        return actuals.clone();
    }

    /**
     * Returns the bipartition threshold. Only one global value is supported.
     * 
     * @return bipartition threshold
     */
    public double getBipartitionThreshold()
    {
        return bipartitionThreshold;
    }

    /**
     * predictions
     */
    double[][] predictions;
    /**
     * gold standard
     */
    int[][] actuals;
    /**
     * bipartition threshold
     */
    double bipartitionThreshold;
    /**
     * instanceIds
     */
    String[] instanceIds;

    public String[] getInstanceIds()
    {
        return instanceIds.clone();
    }

    /**
     * Calculates a bipartition from the predictions (which are usually rankings)
     * 
     * @return a bipartition from the predictions as LxN matrix, using the specified threshold
     */
    public int[][] getPredictionsBipartition()
    {
        int[][] bipartitionMatrix = new int[predictions.length][predictions[0].length];
        for (int i = 0; i < predictions.length; i++) {
            for (int j = 0; j < predictions[i].length; j++) {
                bipartitionMatrix[i][j] = predictions[i][j] >= bipartitionThreshold ? 1 : 0;
            }
        }
        return bipartitionMatrix;
    }

}