ml/regression/JavaLinearRegressionSuite.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.ml.regression;
0019
0020 import java.io.IOException;
0021 import java.util.List;
0022
0023 import org.junit.Test;
0024 import static org.junit.Assert.assertEquals;
0025
0026 import org.apache.spark.SharedSparkSession;
0027 import org.apache.spark.api.java.JavaRDD;
0028 import static org.apache.spark.ml.classification.LogisticRegressionSuite.generateLogisticInputAsList;
0029 import org.apache.spark.ml.feature.LabeledPoint;
0030 import org.apache.spark.sql.Dataset;
0031 import org.apache.spark.sql.Row;
0032
0033 public class JavaLinearRegressionSuite extends SharedSparkSession {
0034   private transient Dataset<Row> dataset;
0035   private transient JavaRDD<LabeledPoint> datasetRDD;
0036
0037   @Override
0038   public void setUp() throws IOException {
0039     super.setUp();
0040     List<LabeledPoint> points = generateLogisticInputAsList(1.0, 1.0, 100, 42);
0041     datasetRDD = jsc.parallelize(points, 2);
0042     dataset = spark.createDataFrame(datasetRDD, LabeledPoint.class);
0043     dataset.createOrReplaceTempView("dataset");
0044   }
0045
0046   @Test
0047   public void linearRegressionDefaultParams() {
0048     LinearRegression lr = new LinearRegression();
0049     assertEquals("label", lr.getLabelCol());
0050     assertEquals("auto", lr.getSolver());
0051     LinearRegressionModel model = lr.fit(dataset);
0052     model.transform(dataset).createOrReplaceTempView("prediction");
0053     Dataset<Row> predictions = spark.sql("SELECT label, prediction FROM prediction");
0054     predictions.collect();
0055     // Check defaults
0056     assertEquals("features", model.getFeaturesCol());
0057     assertEquals("prediction", model.getPredictionCol());
0058   }
0059
0060   @Test
0061   public void linearRegressionWithSetters() {
0062     // Set params, train, and check as many params as we can.
0063     LinearRegression lr = new LinearRegression()
0064       .setMaxIter(10)
0065       .setRegParam(1.0).setSolver("l-bfgs");
0066     LinearRegressionModel model = lr.fit(dataset);
0067     LinearRegression parent = (LinearRegression) model.parent();
0068     assertEquals(10, parent.getMaxIter());
0069     assertEquals(1.0, parent.getRegParam(), 0.0);
0070
0071     // Call fit() with new params, and check as many params as we can.
0072     LinearRegressionModel model2 =
0073       lr.fit(dataset, lr.maxIter().w(5), lr.regParam().w(0.1), lr.predictionCol().w("thePred"));
0074     LinearRegression parent2 = (LinearRegression) model2.parent();
0075     assertEquals(5, parent2.getMaxIter());
0076     assertEquals(0.1, parent2.getRegParam(), 0.0);
0077     assertEquals("thePred", model2.getPredictionCol());
0078   }
0079 }