org.apache.mahout.classifier.sgd.LogisticModelParameters.java Source code

Java tutorial

Introduction

Here is the source code for org.apache.mahout.classifier.sgd.LogisticModelParameters.java

Source

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.mahout.classifier.sgd;

import com.google.common.collect.Maps;
import com.google.gson.*;
import org.apache.mahout.math.DenseMatrix;
import org.apache.mahout.math.Matrix;

import java.io.*;
import java.lang.reflect.Type;
import java.util.Iterator;
import java.util.List;
import java.util.Map;

/**
 * Encapsulates everything we need to know about a model and how it reads and vectorizes its input.
 * This encapsulation allows us to coherently save and restore a model from a file.  This also
 * allows us to keep command line arguments that affect learning in a coherent way.
 */
public class LogisticModelParameters {
    private String targetVariable;
    private Map<String, String> typeMap;
    private int numFeatures;
    private boolean useBias;
    private int maxTargetCategories;
    private List<String> targetCategories = null;
    private double lambda;
    private double learningRate;
    private transient CsvRecordFactory csv = null;
    private OnlineLogisticRegression lr = null;

    public LogisticModelParameters() {
    }

    /**
     * Returns a CsvRecordFactory compatible with this logistic model.  The reason that this is tied
     * in here is so that we have access to the list of target categories when it comes time to save
     * the model.  If the input isn't CSV, then calling setTargetCategories before calling saveTo will
     * suffice.
     *
     * @return
     */
    public CsvRecordFactory getCsvRecordFactory() {
        if (csv == null) {
            csv = new CsvRecordFactory(getTargetVariable(), getTypeMap()).maxTargetValue(getMaxTargetCategories())
                    .includeBiasTerm(useBias());
            if (targetCategories != null) {
                csv.defineTargetCategories(targetCategories);
            }
        }
        return csv;
    }

    /**
     * Creates a logistic regression trainer using the parameters collected here.
     *
     * @return
     */
    public OnlineLogisticRegression createRegression() {
        if (lr == null) {
            lr = new OnlineLogisticRegression(getMaxTargetCategories(), getNumFeatures(), new L1())
                    .lambda(getLambda()).learningRate(getLearningRate()).alpha(1 - 1e-3);
        }
        return lr;
    }

    public static void saveModel(Writer out, OnlineLogisticRegression model, List<String> targetCategories)
            throws IOException {
        LogisticModelParameters x = new LogisticModelParameters();
        x.setTargetCategories(targetCategories);
        x.setLambda(model.getLambda());
        x.setLearningRate(model.currentLearningRate());
        x.setNumFeatures(model.numFeatures());
        x.setUseBias(true);
        x.setTargetCategories(targetCategories);
        x.saveTo(out);
    }

    /**
     * Saves a model in JSON format.  This includes the current state of the logistic regression
     * trainer and the dictionary for the target categories.
     *
     * @param out Where to write the model.
     * @throws IOException
     */
    public void saveTo(Writer out) throws IOException {
        if (lr != null) {
            lr.close();
        }
        targetCategories = csv.getTargetCategories();
        GsonBuilder gb = new GsonBuilder();
        gb.registerTypeAdapter(Matrix.class, new MatrixTypeAdapter());
        Gson gson = gb.setPrettyPrinting().create();

        String savedForm = gson.toJson(this);
        out.write(savedForm);
    }

    /**
     * Reads a model in JSON format.
     *
     * @param in Where to read the model from.
     * @return The LogisticModelParameters object that we read.
     */
    public static LogisticModelParameters loadFrom(Reader in) {
        GsonBuilder gb = new GsonBuilder();
        gb.registerTypeAdapter(Matrix.class, new MatrixTypeAdapter());
        return gb.create().fromJson(in, LogisticModelParameters.class);
    }

    /**
     * Reads a model in JSON format from a File.
     *
     * @param in Where to read the model from.
     * @return The LogisticModelParameters object that we read.
     * @throws IOException If there is an error opening or closing the file.
     */
    public static LogisticModelParameters loadFrom(File in) throws IOException {
        FileReader input = new FileReader(in);
        LogisticModelParameters r = loadFrom(input);
        input.close();
        return r;
    }

    /**
     * Sets the types of the predictors.  This will later be used when reading CSV data.  If you don't
     * use the CSV data and convert to vectors on your own, you don't need to call this.
     *
     * @param predictorList The list of variable names.
     * @param typeList      The list of types in the format preferred by CsvRecordFactory.
     */
    public void setTypeMap(List predictorList, List typeList) {
        typeMap = Maps.newHashMap();
        if (typeList.size() == 0) {
            throw new IllegalArgumentException("Must have at least one type specifier");
        }
        Iterator iTypes = typeList.iterator();
        String lastType = null;
        for (Object x : predictorList) {
            // type list can be short .. we just repeat last spec
            if (iTypes.hasNext()) {
                lastType = iTypes.next().toString();
            }
            typeMap.put(x.toString(), lastType);
        }
    }

    /**
     * Sets the target variable.  If you don't use the CSV record factory, then this is irrelevant.
     *
     * @param targetVariable
     */
    public void setTargetVariable(String targetVariable) {
        this.targetVariable = targetVariable;
    }

    /**
     * Sets the number of target categories to be considered.
     *
     * @param maxTargetCategories
     */
    public void setMaxTargetCategories(int maxTargetCategories) {
        this.maxTargetCategories = maxTargetCategories;
    }

    public void setNumFeatures(int numFeatures) {
        this.numFeatures = numFeatures;
    }

    public void setTargetCategories(List<String> targetCategories) {
        this.targetCategories = targetCategories;
        maxTargetCategories = targetCategories.size();
    }

    public void setUseBias(boolean useBias) {
        this.useBias = useBias;
    }

    public boolean useBias() {
        return useBias;
    }

    public String getTargetVariable() {
        return targetVariable;
    }

    public Map<String, String> getTypeMap() {
        return typeMap;
    }

    public int getNumFeatures() {
        return numFeatures;
    }

    public int getMaxTargetCategories() {
        return maxTargetCategories;
    }

    public double getLambda() {
        return lambda;
    }

    public void setLambda(double lambda) {
        this.lambda = lambda;
    }

    public double getLearningRate() {
        return learningRate;
    }

    public void setLearningRate(double learningRate) {
        this.learningRate = learningRate;
    }

    /**
     * Tells GSON how to (de)serialize a Mahout matrix.  We assume on deserialization that
     * the matrix is dense.
     */
    public static class MatrixTypeAdapter
            implements JsonDeserializer<Matrix>, JsonSerializer<Matrix>, InstanceCreator<Matrix> {
        @Override
        public JsonElement serialize(Matrix m, Type type, JsonSerializationContext jsonSerializationContext) {
            JsonObject r = new JsonObject();
            r.add("rows", new JsonPrimitive(m.numRows()));
            r.add("cols", new JsonPrimitive(m.numCols()));
            JsonArray v = new JsonArray();
            for (int row = 0; row < m.numRows(); row++) {
                JsonArray rowData = new JsonArray();
                for (int col = 0; col < m.numCols(); col++) {
                    rowData.add(new JsonPrimitive(m.get(row, col)));
                }
                v.add(rowData);
            }
            r.add("data", v);
            return r;
        }

        @Override
        public Matrix deserialize(JsonElement x, Type type, JsonDeserializationContext jsonDeserializationContext)
                throws JsonParseException {
            JsonObject data = x.getAsJsonObject();
            Matrix r = new DenseMatrix(data.get("rows").getAsInt(), data.get("cols").getAsInt());
            int i = 0;
            for (JsonElement row : data.get("data").getAsJsonArray()) {
                int j = 0;
                for (JsonElement element : row.getAsJsonArray()) {
                    r.set(i, j, element.getAsDouble());
                    j++;
                }
                i++;
            }
            return r;
        }

        @Override
        public Matrix createInstance(Type type) {
            return new DenseMatrix();
        }
    }
}