sklearn.svm.BaseLibSVMClassifier.java Source code

Introduction

Here is the source code for sklearn.svm.BaseLibSVMClassifier.java

Source

/*
 * Copyright (c) 2015 Villu Ruusmann
 *
 * This file is part of JPMML-SkLearn
 *
 * JPMML-SkLearn is free software: you can redistribute it and/or modify
 * it under the terms of the GNU Affero General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * JPMML-SkLearn 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 Affero General Public License for more details.
 *
 * You should have received a copy of the GNU Affero General Public License
 * along with JPMML-SkLearn.  If not, see <http://www.gnu.org/licenses/>.
 */
package sklearn.svm;

import java.util.ArrayList;
import java.util.List;

import com.google.common.collect.Iterables;
import org.dmg.pmml.MiningFunction;
import org.dmg.pmml.support_vector_machine.Kernel;
import org.dmg.pmml.support_vector_machine.SupportVectorMachine;
import org.dmg.pmml.support_vector_machine.SupportVectorMachineModel;
import org.dmg.pmml.support_vector_machine.VectorDictionary;
import org.dmg.pmml.support_vector_machine.VectorInstance;
import org.jpmml.converter.ModelUtil;
import org.jpmml.converter.Schema;
import org.jpmml.converter.ValueUtil;
import org.jpmml.sklearn.ClassDictUtil;
import org.jpmml.sklearn.MatrixUtil;
import sklearn.Classifier;

abstract public class BaseLibSVMClassifier extends Classifier {

    public BaseLibSVMClassifier(String module, String name) {
        super(module, name);
    }

    @Override
    public int getNumberOfFeatures() {
        int[] shape = getSupportVectorsShape();

        return shape[1];
    }

    @Override
    public boolean requiresContinuousInput() {
        return false;
    }

    @Override
    public boolean hasProbabilityDistribution() {
        return false;
    }

    @Override
    public SupportVectorMachineModel encodeModel(Schema schema) {
        int[] shape = getSupportVectorsShape();

        int numberOfVectors = shape[0];
        int numberOfFeatures = shape[1];

        List<Integer> support = getSupport();
        List<? extends Number> supportVectors = getSupportVectors();
        List<Integer> supportSizes = getSupportSizes();
        List<? extends Number> dualCoef = getDualCoef();
        List<? extends Number> intercept = getIntercept();

        int[] offsets = new int[supportSizes.size() + 1];

        for (int i = 0; i < supportSizes.size(); i++) {
            offsets[i + 1] = offsets[i] + supportSizes.get(i);
        }

        VectorDictionary vectorDictionary = SupportVectorMachineUtil.encodeVectorDictionary(support, supportVectors,
                numberOfVectors, numberOfFeatures, schema);

        List<VectorInstance> vectorInstances = vectorDictionary.getVectorInstances();

        Kernel kernel = SupportVectorMachineUtil.encodeKernel(getKernel(), getDegree(), getGamma(), getCoef0());

        List<SupportVectorMachine> supportVectorMachines = new ArrayList<>();

        int i = 0;

        List<String> targetCategories = schema.getTargetCategories();

        for (int first = 0; first < targetCategories.size(); first++) {

            for (int second = first + 1; second < targetCategories.size(); second++) {
                List<VectorInstance> svmVectorInstances = new ArrayList<>();
                svmVectorInstances.addAll(slice(vectorInstances, offsets, first));
                svmVectorInstances.addAll(slice(vectorInstances, offsets, second));

                List<Number> svmDualCoef = new ArrayList<>();
                svmDualCoef.addAll(
                        slice(MatrixUtil.getRow(dualCoef, targetCategories.size() - 1, numberOfVectors, second - 1),
                                offsets, first));
                svmDualCoef.addAll(
                        slice(MatrixUtil.getRow(dualCoef, targetCategories.size() - 1, numberOfVectors, first),
                                offsets, second));

                // LibSVM: (decisionFunction > 0 ? first : second)
                // PMML: (decisionFunction < 0 ? first : second)
                SupportVectorMachine supportVectorMachine = SupportVectorMachineUtil
                        .encodeSupportVectorMachine(svmVectorInstances, svmDualCoef, Iterables.get(intercept, i))
                        .setTargetCategory(targetCategories.get(second))
                        .setAlternateTargetCategory(targetCategories.get(first));

                supportVectorMachines.add(supportVectorMachine);

                i++;
            }
        }

        SupportVectorMachineModel supportVectorMachineModel = new SupportVectorMachineModel(
                MiningFunction.CLASSIFICATION, ModelUtil.createMiningSchema(schema), vectorDictionary,
                supportVectorMachines)
                        .setClassificationMethod(SupportVectorMachineModel.ClassificationMethod.ONE_AGAINST_ONE)
                        .setKernel(kernel);

        return supportVectorMachineModel;
    }

    public String getKernel() {
        return (String) get("kernel");
    }

    public Integer getDegree() {
        return ValueUtil.asInteger((Number) get("degree"));
    }

    public Double getGamma() {
        return ValueUtil.asDouble((Number) get("_gamma"));
    }

    public Double getCoef0() {
        return ValueUtil.asDouble((Number) get("coef0"));
    }

    public List<Integer> getSupport() {
        return ValueUtil.asIntegers((List) ClassDictUtil.getArray(this, "support_"));
    }

    public List<? extends Number> getSupportVectors() {
        return (List) ClassDictUtil.getArray(this, "support_vectors_");
    }

    public List<Integer> getSupportSizes() {
        return ValueUtil.asIntegers((List) ClassDictUtil.getArray(this, "n_support_"));
    }

    public List<? extends Number> getDualCoef() {
        return (List) ClassDictUtil.getArray(this, "_dual_coef_");
    }

    public List<? extends Number> getIntercept() {
        return (List) ClassDictUtil.getArray(this, "_intercept_");
    }

    private int[] getSupportVectorsShape() {
        return ClassDictUtil.getShape(this, "support_vectors_", 2);
    }

    static private <E> List<E> slice(List<E> list, int[] offsets, int index) {
        return list.subList(offsets[index], offsets[index + 1]);
    }
}