examples/ml/JavaRandomForestClassifierExample.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.examples.ml;
0019
0020 // $example on$
0021 import org.apache.spark.ml.Pipeline;
0022 import org.apache.spark.ml.PipelineModel;
0023 import org.apache.spark.ml.PipelineStage;
0024 import org.apache.spark.ml.classification.RandomForestClassificationModel;
0025 import org.apache.spark.ml.classification.RandomForestClassifier;
0026 import org.apache.spark.ml.evaluation.MulticlassClassificationEvaluator;
0027 import org.apache.spark.ml.feature.*;
0028 import org.apache.spark.sql.Dataset;
0029 import org.apache.spark.sql.Row;
0030 import org.apache.spark.sql.SparkSession;
0031 // $example off$
0032
0033 public class JavaRandomForestClassifierExample {
0034   public static void main(String[] args) {
0035     SparkSession spark = SparkSession
0036       .builder()
0037       .appName("JavaRandomForestClassifierExample")
0038       .getOrCreate();
0039
0040     // $example on$
0041     // Load and parse the data file, converting it to a DataFrame.
0042     Dataset<Row> data = spark.read().format("libsvm").load("data/mllib/sample_libsvm_data.txt");
0043
0044     // Index labels, adding metadata to the label column.
0045     // Fit on whole dataset to include all labels in index.
0046     StringIndexerModel labelIndexer = new StringIndexer()
0047       .setInputCol("label")
0048       .setOutputCol("indexedLabel")
0049       .fit(data);
0050     // Automatically identify categorical features, and index them.
0051     // Set maxCategories so features with > 4 distinct values are treated as continuous.
0052     VectorIndexerModel featureIndexer = new VectorIndexer()
0053       .setInputCol("features")
0054       .setOutputCol("indexedFeatures")
0055       .setMaxCategories(4)
0056       .fit(data);
0057
0058     // Split the data into training and test sets (30% held out for testing)
0059     Dataset<Row>[] splits = data.randomSplit(new double[] {0.7, 0.3});
0060     Dataset<Row> trainingData = splits[0];
0061     Dataset<Row> testData = splits[1];
0062
0063     // Train a RandomForest model.
0064     RandomForestClassifier rf = new RandomForestClassifier()
0065       .setLabelCol("indexedLabel")
0066       .setFeaturesCol("indexedFeatures");
0067
0068     // Convert indexed labels back to original labels.
0069     IndexToString labelConverter = new IndexToString()
0070       .setInputCol("prediction")
0071       .setOutputCol("predictedLabel")
0072       .setLabels(labelIndexer.labelsArray()[0]);
0073
0074     // Chain indexers and forest in a Pipeline
0075     Pipeline pipeline = new Pipeline()
0076       .setStages(new PipelineStage[] {labelIndexer, featureIndexer, rf, labelConverter});
0077
0078     // Train model. This also runs the indexers.
0079     PipelineModel model = pipeline.fit(trainingData);
0080
0081     // Make predictions.
0082     Dataset<Row> predictions = model.transform(testData);
0083
0084     // Select example rows to display.
0085     predictions.select("predictedLabel", "label", "features").show(5);
0086
0087     // Select (prediction, true label) and compute test error
0088     MulticlassClassificationEvaluator evaluator = new MulticlassClassificationEvaluator()
0089       .setLabelCol("indexedLabel")
0090       .setPredictionCol("prediction")
0091       .setMetricName("accuracy");
0092     double accuracy = evaluator.evaluate(predictions);
0093     System.out.println("Test Error = " + (1.0 - accuracy));
0094
0095     RandomForestClassificationModel rfModel = (RandomForestClassificationModel)(model.stages()[2]);
0096     System.out.println("Learned classification forest model:\n" + rfModel.toDebugString());
0097     // $example off$
0098
0099     spark.stop();
0100   }
0101 }