mllib/classification/JavaSVMSuite.java

0001 /*
0002  * Licensed to the Apache Software Foundation (ASF) under one or more
0003  * contributor license agreements.  See the NOTICE file distributed with
0004  * this work for additional information regarding copyright ownership.
0005  * The ASF licenses this file to You under the Apache License, Version 2.0
0006  * (the "License"); you may not use this file except in compliance with
0007  * the License.  You may obtain a copy of the License at
0008  *
0009  *    http://www.apache.org/licenses/LICENSE-2.0
0010  *
0011  * Unless required by applicable law or agreed to in writing, software
0012  * distributed under the License is distributed on an "AS IS" BASIS,
0013  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
0014  * See the License for the specific language governing permissions and
0015  * limitations under the License.
0016  */
0017
0018 package org.apache.spark.mllib.classification;
0019
0020 import java.util.List;
0021
0022 import org.junit.Assert;
0023 import org.junit.Test;
0024
0025 import org.apache.spark.SharedSparkSession;
0026 import org.apache.spark.api.java.JavaRDD;
0027 import org.apache.spark.mllib.regression.LabeledPoint;
0028
0029 public class JavaSVMSuite extends SharedSparkSession {
0030
0031   int validatePrediction(List<LabeledPoint> validationData, SVMModel model) {
0032     int numAccurate = 0;
0033     for (LabeledPoint point : validationData) {
0034       Double prediction = model.predict(point.features());
0035       if (prediction == point.label()) {
0036         numAccurate++;
0037       }
0038     }
0039     return numAccurate;
0040   }
0041
0042   @Test
0043   public void runSVMUsingConstructor() {
0044     int nPoints = 10000;
0045     double A = 2.0;
0046     double[] weights = {-1.5, 1.0};
0047
0048     JavaRDD<LabeledPoint> testRDD = jsc.parallelize(SVMSuite.generateSVMInputAsList(A,
0049       weights, nPoints, 42), 2).cache();
0050     List<LabeledPoint> validationData =
0051       SVMSuite.generateSVMInputAsList(A, weights, nPoints, 17);
0052
0053     SVMWithSGD svmSGDImpl = new SVMWithSGD();
0054     svmSGDImpl.setIntercept(true);
0055     svmSGDImpl.optimizer().setStepSize(1.0)
0056       .setRegParam(1.0)
0057       .setNumIterations(100);
0058     SVMModel model = svmSGDImpl.run(testRDD.rdd());
0059
0060     int numAccurate = validatePrediction(validationData, model);
0061     Assert.assertTrue(numAccurate > nPoints * 4.0 / 5.0);
0062   }
0063
0064   @Test
0065   public void runSVMUsingStaticMethods() {
0066     int nPoints = 10000;
0067     double A = 0.0;
0068     double[] weights = {-1.5, 1.0};
0069
0070     JavaRDD<LabeledPoint> testRDD = jsc.parallelize(SVMSuite.generateSVMInputAsList(A,
0071       weights, nPoints, 42), 2).cache();
0072     List<LabeledPoint> validationData =
0073       SVMSuite.generateSVMInputAsList(A, weights, nPoints, 17);
0074
0075     SVMModel model = SVMWithSGD.train(testRDD.rdd(), 100, 1.0, 1.0, 1.0);
0076
0077     int numAccurate = validatePrediction(validationData, model);
0078     Assert.assertTrue(numAccurate > nPoints * 4.0 / 5.0);
0079   }
0080 }