spark MLlib Classification and regression 学习

二分类：SVMs，logistic regression，decision trees，random forests，gradient-boosted trees，naive Bayes

多分类： logistic regression，decision trees，random forests， naive Bayes

归回： linear least regression， decision tress，random forests，gradient-boosted trees， isotonic regression。

一。Linear models

　　classification (SVMs, logistic regression)

 package ML.ClassificationAndRegression;

 import org.apache.spark.SparkConf;

 import org.apache.spark.api.java.JavaRDD;

 import org.apache.spark.api.java.JavaSparkContext;

 import org.apache.spark.api.java.function.Function;

 import org.apache.spark.mllib.classification.SVMModel;

 import org.apache.spark.mllib.classification.SVMWithSGD;

 import org.apache.spark.mllib.evaluation.BinaryClassificationMetrics;

 import org.apache.spark.mllib.optimization.L1Updater;

 import org.apache.spark.mllib.regression.LabeledPoint;

 import org.apache.spark.mllib.util.MLUtils;

 import org.apache.spark.rdd.RDD;

 import scala.Tuple2;

 /**

  * TODO

  *

  * @ClassName: SVMClassifier

  * @author: DingH

  * @since: 2019/4/9 10:28

  */

 public class SVMClassifier {

     public static void main(String[] args) {

         SparkConf conf = new SparkConf().setAppName("SVM Classifier Example").setMaster("local");

         JavaSparkContext jsc = new JavaSparkContext(conf);

         String path = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_libsvm_data.txt";

         JavaRDD<LabeledPoint> data = MLUtils.loadLibSVMFile(jsc.sc(), path).toJavaRDD();

         // Split initial RDD into two... [60% training data, 40% testing data].

         JavaRDD<LabeledPoint> train = data.sample(false, 0.6, 11L);

         train.cache();

         final JavaRDD<LabeledPoint> test = data.subtract(train);

         //Run training algorithm to build the model

         int numsIterations = 100;

         SVMWithSGD svm = new SVMWithSGD();

         svm.optimizer().setNumIterations(200).setRegParam(0.01).setUpdater(new L1Updater());

         final SVMModel model1 = svm.run(train.rdd());

 //        final SVMModel model1 = SVMWithSGD.train(train.rdd(), numsIterations);

         model1.clearThreshold();

         JavaRDD<Tuple2<Object, Object>> scoraAndLables = test.map(new Function<LabeledPoint, Tuple2<Object, Object>>() {

             public Tuple2<Object, Object> call(LabeledPoint p) throws Exception {

                 double predict = model1.predict(p.features());

                 return new Tuple2<Object, Object>(predict, p.label());

             }

         });

         BinaryClassificationMetrics metrics = new BinaryClassificationMetrics(scoraAndLables.rdd());

         double areaUnderROC = metrics.areaUnderROC();

         System.out.println("Area under ROC = " + areaUnderROC);

         model1.save(jsc.sc(),"D:\\IdeaProjects\\SimpleApp\\src\\main\\java\\MLModel");

         SVMModel model = SVMModel.load(jsc.sc(), "D:\\IdeaProjects\\SimpleApp\\src\\main\\java\\MLModel");

     }

 }

SVMClassifier

 package ML.ClassificationAndRegression;

 import org.apache.spark.SparkConf;

 import org.apache.spark.api.java.JavaRDD;

 import org.apache.spark.api.java.JavaSparkContext;

 import org.apache.spark.api.java.function.Function;

 import org.apache.spark.ml.classification.MultiClassSummarizer;

 import org.apache.spark.mllib.classification.LogisticRegressionModel;

 import org.apache.spark.mllib.classification.LogisticRegressionWithLBFGS;

 import org.apache.spark.mllib.classification.LogisticRegressionWithSGD;

 import org.apache.spark.mllib.evaluation.MulticlassMetrics;

 import org.apache.spark.mllib.regression.LabeledPoint;

 import org.apache.spark.mllib.util.MLUtils;

 import scala.Tuple2;

 /**

  * TODO

  *

  * @ClassName: LogistiRegression

  * @author: DingH

  * @since: 2019/4/9 11:08

  */

 public class LogistiRegression {

     public static void main(String[] args) {

         SparkConf conf = new SparkConf().setMaster("local").setAppName("LogisticRegression");

         JavaSparkContext jsc = new JavaSparkContext(conf);

         String path = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_libsvm_data.txt";

         JavaRDD<LabeledPoint> data = MLUtils.loadLibSVMFile(jsc.sc(), path).toJavaRDD();

         JavaRDD<LabeledPoint>[] split = data.randomSplit(new double[]{0.6, 0.4}, 11L);

         JavaRDD<LabeledPoint> training = split[0].cache();

         final JavaRDD<LabeledPoint> test = split[1];

         final LogisticRegressionModel model = new LogisticRegressionWithLBFGS().setNumClasses(10).run(training.rdd());

         JavaRDD<Tuple2<Object, Object>> predictionAndLabels = test.map(new Function<LabeledPoint, Tuple2<Object, Object>>() {

             public Tuple2<Object, Object> call(LabeledPoint labeledPoint) throws Exception {

                 double predict = model.predict(labeledPoint.features());

                 return new Tuple2<Object, Object>(predict, labeledPoint.label());

             }

         });

         MulticlassMetrics metrics = new MulticlassMetrics(predictionAndLabels.rdd());

         double precision = metrics.precision();

         System.out.println("Precision = " + precision);

         // Save and load model

 //        model.save(jsc.sc(), "myModelPath");

 //        LogisticRegressionModel sameModel = LogisticRegressionModel.load(jsc.sc(), "myModelPath");

     }

 }

LogistiRegression

　　linear regression (least squares, Lasso, ridge)

 package ML.ClassificationAndRegression;

 import org.apache.hadoop.yarn.webapp.hamlet.Hamlet;

 import org.apache.spark.SparkConf;

 import org.apache.spark.api.java.JavaDoubleRDD;

 import org.apache.spark.api.java.JavaRDD;

 import org.apache.spark.api.java.JavaSparkContext;

 import org.apache.spark.api.java.function.Function;

 import org.apache.spark.mllib.linalg.Vectors;

 import org.apache.spark.mllib.regression.LabeledPoint;

 import org.apache.spark.mllib.regression.LinearRegressionModel;

 import org.apache.spark.mllib.regression.LinearRegressionWithSGD;

 import org.apache.spark.mllib.util.MLUtils;

 import scala.Tuple2;

 /**

  * TODO

  *

  * @ClassName: Regression

  * @author: DingH

  * @since: 2019/4/9 11:21

  */

 public class Regression {

     public static void main(String[] args) {

         SparkConf conf = new SparkConf().setAppName("Regression").setMaster("local");

         JavaSparkContext jsc = new JavaSparkContext(conf);

         String path = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\ridge-data\\lpsa.data";

         JavaRDD<String> data = jsc.textFile(path);

         JavaRDD<LabeledPoint> parsedData = data.map(new Function<String, LabeledPoint>() {

             public LabeledPoint call(String line) throws Exception {

                 String[] split = line.split(",");

                 String[] features = split[1].split(" ");

                 double[] v = new double[features.length];

                 for (int i = 0; i < features.length - 1; i++) {

                     v[i] = Double.parseDouble(features[i]);

                 }

                 return new LabeledPoint(Double.parseDouble(split[0]), Vectors.dense(v));

             }

         }).cache();

         final LinearRegressionModel model = LinearRegressionWithSGD.train(parsedData.rdd(), 100);

         JavaRDD<Tuple2<Double, Double>> valuesAndLabels = parsedData.map(new Function<LabeledPoint, Tuple2<Double, Double>>() {

             public Tuple2<Double, Double> call(LabeledPoint labeledPoint) throws Exception {

                 double predict = model.predict(labeledPoint.features());

                 return new Tuple2<Double, Double>(predict, labeledPoint.label());

             }

         });

         Double MSE = new JavaDoubleRDD(valuesAndLabels.map(

                 new Function<Tuple2<Double, Double>, Object>() {

                     public Object call(Tuple2<Double, Double> dat) throws Exception {

                         return Math.pow(dat._1 - dat._2, 2.0);

                     }

                 }

         ).rdd()).mean();

         System.out.println("training Mean Squared Error = " + MSE);

         // Save and load model

 //        model.save(jsc.sc(), "myModelPath");

 //        LinearRegressionModel sameModel = LinearRegressionModel.load(jsc.sc(), "myModelPath");

     }

 }

Regression

二。Decision Trees.

problem specification parameters: algo, numClasses, categoricalFeaturesInfo

stopping criteria : maxDepth, minInfoGain, minInstancePerNode

tunnable parameters: maxBins, impurity,

package ML.DT;

import org.apache.spark.SparkConf;

import org.apache.spark.api.java.JavaPairRDD;

import org.apache.spark.api.java.JavaRDD;

import org.apache.spark.api.java.JavaSparkContext;

import org.apache.spark.api.java.function.Function;

import org.apache.spark.api.java.function.PairFunction;

import org.apache.spark.mllib.regression.LabeledPoint;

import org.apache.spark.mllib.tree.DecisionTree;

import org.apache.spark.mllib.tree.model.DecisionTreeModel;

import org.apache.spark.mllib.util.MLUtils;

import scala.Tuple2;

import java.util.HashMap;

/**

 * TODO

 *

 * @ClassName: classification

 * @author: DingH

 * @since: 2019/4/9 16:11

 */

public class classification {

    public static void main(String[] args) {

        SparkConf conf = new SparkConf().setAppName("DTclassification").setMaster("local");

        JavaSparkContext jsc = new JavaSparkContext(conf);

        String path = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_libsvm_data.txt";

        JavaRDD<LabeledPoint> data = MLUtils.loadLibSVMFile(jsc.sc(), path).toJavaRDD();

        JavaRDD<LabeledPoint>[] split = data.randomSplit(new double[]{0.7, 0.3}, 11L);

        JavaRDD<LabeledPoint> trainningData = split[0];

        JavaRDD<LabeledPoint> test = split[1];

        int numsClasses = 2;

        HashMap<Integer, Integer> categoricalFeaturesInfo  = new HashMap<Integer, Integer>();

        String impurity = "gini";

        int maxDepth  = 1;

        int maxbins = 32;

        final DecisionTreeModel model = DecisionTree.trainClassifier(trainningData, numsClasses,categoricalFeaturesInfo, impurity, maxDepth,maxbins);

        JavaPairRDD<Double, Double> predictionAndLable = test.mapToPair(new PairFunction<LabeledPoint, Double, Double>() {

            public Tuple2<Double, Double> call(LabeledPoint labeledPoint) throws Exception {

                return new Tuple2<Double, Double>(model.predict(labeledPoint.features()), labeledPoint.label());

            }

        });

        double testErr = predictionAndLable.filter(new Function<Tuple2<Double, Double>, Boolean>() {

            public Boolean call(Tuple2<Double, Double> doubleDoubleTuple2) throws Exception {

                return !doubleDoubleTuple2._1().equals(doubleDoubleTuple2._2());

            }

        }).count() * 1.0 / test.count();

        System.out.println("Test Error: " + testErr);

        System.out.println("Learned classification tree model:\n" + model.toDebugString());

        // Save and load model

//        model.save(jsc.sc(), "target/tmp/myDecisionTreeClassificationModel");

//        DecisionTreeModel sameModel = DecisionTreeModel.load(jsc.sc(), "target/tmp/myDecisionTreeClassificationModel");

    }

}

classification

package ML.DT;

import org.apache.spark.SparkConf;

import org.apache.spark.api.java.JavaPairRDD;

import org.apache.spark.api.java.JavaRDD;

import org.apache.spark.api.java.JavaSparkContext;

import org.apache.spark.api.java.function.Function;

import org.apache.spark.api.java.function.Function2;

import org.apache.spark.api.java.function.PairFunction;

import org.apache.spark.mllib.regression.LabeledPoint;

import org.apache.spark.mllib.tree.DecisionTree;

import org.apache.spark.mllib.tree.model.DecisionTreeModel;

import org.apache.spark.mllib.util.MLUtils;

import scala.Tuple2;

import java.util.HashMap;

import java.util.Map;

/**

 * TODO

 *

 * @ClassName: Regression

 * @author: DingH

 * @since: 2019/4/9 16:33

 */

public class Regression {

    public static void main(String[] args) {

        SparkConf sparkConf = new SparkConf().setAppName("JavaDecisionTreeRegressionExample").setMaster("local");

        JavaSparkContext jsc = new JavaSparkContext(sparkConf);

        // Load and parse the data file.

        String datapath = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_libsvm_data.txt";

        JavaRDD<LabeledPoint> data = MLUtils.loadLibSVMFile(jsc.sc(), datapath).toJavaRDD();

        // Split the data into training and test sets (30% held out for testing)

        JavaRDD<LabeledPoint>[] splits = data.randomSplit(new double[]{0.7, 0.3});

        JavaRDD<LabeledPoint> trainingData = splits[0];

        JavaRDD<LabeledPoint> testData = splits[1];

        // Set parameters.

        // Empty categoricalFeaturesInfo indicates all features are continuous.

        Map<Integer, Integer> categoricalFeaturesInfo = new HashMap<Integer, Integer>();

        String impurity = "variance";

        Integer maxDepth = 5;

        Integer maxBins = 32;

        // Train a DecisionTree model.

        final DecisionTreeModel model = DecisionTree.trainRegressor(trainingData, categoricalFeaturesInfo, impurity, maxDepth, maxBins);

        // Evaluate model on test instances and compute test error

        JavaPairRDD<Double, Double> predictionAndLabel = testData.mapToPair(new PairFunction<LabeledPoint, Double, Double>() {

            public Tuple2<Double, Double> call(LabeledPoint p) {

                return new Tuple2<Double, Double>(model.predict(p.features()), p.label());

            }

        });

        Double testMSE = predictionAndLabel.map(new Function<Tuple2<Double, Double>, Double>() {

            public Double call(Tuple2<Double, Double> pl) {

                Double diff = pl._1() - pl._2();

                return diff * diff;

            }

        }).reduce(new Function2<Double, Double, Double>() {

            public Double call(Double a, Double b) {

                return a + b;

            }

        }) / data.count();

        System.out.println("Test Mean Squared Error: " + testMSE);

        System.out.println("Learned regression tree model:\n" + model.toDebugString());

        // Save and load model

//        model.save(jsc.sc(), "target/tmp/myDecisionTreeRegressionModel");

//        DecisionTreeModel sameModel = DecisionTreeModel.load(jsc.sc(), "target/tmp/myDecisionTreeRegressionModel");

    }

}

Regression

三。Random Forests

　　样本随机，特征随机

　　featureSubsetStrategy - Number of features to consider for splits at each node. Supported: "auto", "all", "sqrt", "log2", "onethird". If "auto" is set, this parameter is set based on numTrees: if numTrees == 1, set to "all"; if numTrees > 1 (forest) set to "sqrt".

package ML.RF;

import org.apache.spark.SparkConf;

import org.apache.spark.api.java.JavaPairRDD;

import org.apache.spark.api.java.JavaRDD;

import org.apache.spark.api.java.JavaSparkContext;

import org.apache.spark.api.java.function.Function;

import org.apache.spark.api.java.function.PairFunction;

import org.apache.spark.mllib.regression.LabeledPoint;

import org.apache.spark.mllib.tree.RandomForest;

import org.apache.spark.mllib.tree.model.RandomForestModel;

import org.apache.spark.mllib.util.MLUtils;

import scala.Tuple2;

import java.util.HashMap;

/**

 * TODO

 *

 * @ClassName: classification

 * @author: DingH

 * @since: 2019/4/9 16:58

 */

public class classification {

    public static void main(String[] args) {

        SparkConf sparkConf = new SparkConf().setMaster("local").setAppName("JavaRandomForestClassificationExample");

        JavaSparkContext jsc = new JavaSparkContext(sparkConf);

        // Load and parse the data file.

        String datapath = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_libsvm_data.txt";

        JavaRDD<LabeledPoint> data = MLUtils.loadLibSVMFile(jsc.sc(), datapath).toJavaRDD();

        // Split the data into training and test sets (30% held out for testing)

        JavaRDD<LabeledPoint>[] splits = data.randomSplit(new double[]{0.7, 0.3});

        JavaRDD<LabeledPoint> trainingData = splits[0];

        JavaRDD<LabeledPoint> testData = splits[1];

        // Train a RandomForest model.

        // Empty categoricalFeaturesInfo indicates all features are continuous.

        Integer numClasses = 2;

        HashMap<Integer, Integer> categoricalFeaturesInfo = new HashMap<Integer, Integer>();

        Integer numTrees = 3; // Use more in practice.

        String featureSubsetStrategy = "auto"; // Let the algorithm choose.

        String impurity = "gini";

        Integer maxDepth = 5;

        Integer maxBins = 32;

        Integer seed = 12345;

        final RandomForestModel model = RandomForest.trainClassifier(trainingData, numClasses,categoricalFeaturesInfo, numTrees, featureSubsetStrategy, impurity, maxDepth, maxBins, seed);

        // Evaluate model on test instances and compute test error

        JavaPairRDD<Double, Double> predictionAndLabel = testData.mapToPair(new PairFunction<LabeledPoint, Double, Double>() {

            public Tuple2<Double, Double> call(LabeledPoint p) {

              return new Tuple2<Double, Double>(model.predict(p.features()), p.label());

            }

          });

        Double testErr =

          1.0 * predictionAndLabel.filter(new Function<Tuple2<Double, Double>, Boolean>() {

            public Boolean call(Tuple2<Double, Double> pl) {

              return !pl._1().equals(pl._2());

            }

          }).count() / testData.count();

        System.out.println("Test Error: " + testErr);

        System.out.println("Learned classification forest model:\n" + model.toDebugString());

        // Save and load model

//        model.save(jsc.sc(), "target/tmp/myRandomForestClassificationModel");

//        RandomForestModel sameModel = RandomForestModel.load(jsc.sc(),"target/tmp/myRandomForestClassificationModel");

    }

}

classification

package ML.RF;

import org.apache.spark.SparkConf;

import org.apache.spark.api.java.JavaPairRDD;

import org.apache.spark.api.java.JavaRDD;

import org.apache.spark.api.java.JavaSparkContext;

import org.apache.spark.api.java.function.Function;

import org.apache.spark.api.java.function.Function2;

import org.apache.spark.api.java.function.PairFunction;

import org.apache.spark.mllib.regression.LabeledPoint;

import org.apache.spark.mllib.tree.RandomForest;

import org.apache.spark.mllib.tree.model.RandomForestModel;

import org.apache.spark.mllib.util.MLUtils;

import scala.Tuple2;

import java.util.HashMap;

import java.util.Map;

/**

 * TODO

 *

 * @ClassName: regression

 * @author: DingH

 * @since: 2019/4/9 17:50

 */

public class regression {

    public static void main(String[] args) {

        SparkConf sparkConf = new SparkConf().setMaster("local").setAppName("JavaRandomForestRegressionExample");

        JavaSparkContext jsc = new JavaSparkContext(sparkConf);

        // Load and parse the data file.

        String datapath = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_libsvm_data.txt";

        JavaRDD<LabeledPoint> data = MLUtils.loadLibSVMFile(jsc.sc(), datapath).toJavaRDD();

        // Split the data into training and test sets (30% held out for testing)

        JavaRDD<LabeledPoint>[] splits = data.randomSplit(new double[]{0.7, 0.3});

        JavaRDD<LabeledPoint> trainingData = splits[0];

        JavaRDD<LabeledPoint> testData = splits[1];

        // Set parameters.

        // Empty categoricalFeaturesInfo indicates all features are continuous.

        Map<Integer, Integer> categoricalFeaturesInfo = new HashMap<Integer, Integer>();

        Integer numTrees = 3; // Use more in practice.

        String featureSubsetStrategy = "auto"; // Let the algorithm choose.

        String impurity = "variance";

        Integer maxDepth = 4;

        Integer maxBins = 32;

        Integer seed = 12345;

        // Train a RandomForest model.

        final RandomForestModel model = RandomForest.trainRegressor(trainingData,categoricalFeaturesInfo, numTrees, featureSubsetStrategy, impurity, maxDepth, maxBins, seed);

        // Evaluate model on test instances and compute test error

        JavaPairRDD<Double, Double> predictionAndLabel = testData.mapToPair(new PairFunction<LabeledPoint, Double, Double>() {

            public Tuple2<Double, Double> call(LabeledPoint p) {

              return new Tuple2<Double, Double>(model.predict(p.features()), p.label());

            }

          });

        Double testMSE = predictionAndLabel.map(new Function<Tuple2<Double, Double>, Double>() {

            public Double call(Tuple2<Double, Double> pl) {

              Double diff = pl._1() - pl._2();

              return diff * diff;

            }

          }).reduce(new Function2<Double, Double, Double>() {

            public Double call(Double a, Double b) {

              return a + b;

            }

          }) / testData.count();

        System.out.println("Test Mean Squared Error: " + testMSE);

        System.out.println("Learned regression forest model:\n" + model.toDebugString());

        // Save and load model

        model.save(jsc.sc(), "target/tmp/myRandomForestRegressionModel");

        RandomForestModel sameModel = RandomForestModel.load(jsc.sc(),

          "target/tmp/myRandomForestRegressionModel");

    }

}

regression

四。Gradient-Boosted Trees

Usage tips: loss, numIterations, learningRate, algo

BoostingStrategy.validationTol

package ML.GradientBoostedTrees;

import org.apache.spark.SparkConf;

import org.apache.spark.api.java.JavaPairRDD;

import org.apache.spark.api.java.JavaRDD;

import org.apache.spark.api.java.JavaSparkContext;

import org.apache.spark.api.java.function.Function;

import org.apache.spark.api.java.function.PairFunction;

import org.apache.spark.mllib.regression.LabeledPoint;

import org.apache.spark.mllib.tree.GradientBoostedTrees;

import org.apache.spark.mllib.tree.configuration.BoostingStrategy;

import org.apache.spark.mllib.tree.model.GradientBoostedTreesModel;

import org.apache.spark.mllib.util.MLUtils;

import scala.Tuple2;

import java.util.HashMap;

import java.util.Map;

/**

 * TODO

 *

 * @ClassName: classification

 * @author: DingH

 * @since: 2019/4/9 17:56

 */

public class classification {

    public static void main(String[] args) {

        SparkConf sparkConf = new SparkConf().setMaster("local").setAppName("JavaGradientBoostedTreesClassificationExample");

        JavaSparkContext jsc = new JavaSparkContext(sparkConf);

        // Load and parse the data file.

        String datapath = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_libsvm_data.txt";

        JavaRDD<LabeledPoint> data = MLUtils.loadLibSVMFile(jsc.sc(), datapath).toJavaRDD();

        // Split the data into training and test sets (30% held out for testing)

        JavaRDD<LabeledPoint>[] splits = data.randomSplit(new double[]{0.7, 0.3});

        JavaRDD<LabeledPoint> trainingData = splits[0];

        JavaRDD<LabeledPoint> testData = splits[1];

        // Train a GradientBoostedTrees model.

        // The defaultParams for Classification use LogLoss by default.

        BoostingStrategy boostingStrategy = BoostingStrategy.defaultParams("Classification");

        boostingStrategy.setNumIterations(3); // Note: Use more iterations in practice.

        boostingStrategy.getTreeStrategy().setNumClasses(2);

        boostingStrategy.getTreeStrategy().setMaxDepth(5);

        // Empty categoricalFeaturesInfo indicates all features are continuous.

        Map<Integer, Integer> categoricalFeaturesInfo = new HashMap<Integer, Integer>();

        boostingStrategy.treeStrategy().setCategoricalFeaturesInfo(categoricalFeaturesInfo);

        final GradientBoostedTreesModel model =

          GradientBoostedTrees.train(trainingData, boostingStrategy);

        // Evaluate model on test instances and compute test error

        JavaPairRDD<Double, Double> predictionAndLabel =

          testData.mapToPair(new PairFunction<LabeledPoint, Double, Double>() {

            public Tuple2<Double, Double> call(LabeledPoint p) {

              return new Tuple2<Double, Double>(model.predict(p.features()), p.label());

            }

          });

        Double testErr =

          1.0 * predictionAndLabel.filter(new Function<Tuple2<Double, Double>, Boolean>() {

            public Boolean call(Tuple2<Double, Double> pl) {

              return !pl._1().equals(pl._2());

            }

          }).count() / testData.count();

        System.out.println("Test Error: " + testErr);

        System.out.println("Learned classification GBT model:\n" + model.toDebugString());

        // Save and load model

//        model.save(jsc.sc(), "target/tmp/myGradientBoostingClassificationModel");

//        GradientBoostedTreesModel sameModel = GradientBoostedTreesModel.load(jsc.sc(),

//          "target/tmp/myGradientBoostingClassificationModel");

    }

}

classification

regression

五。naive Bayes

　　model type : "multinomial","bernouli"

 package ML.ClassificationAndRegression.NaiveBayes;

 import org.apache.spark.SparkConf;

 import org.apache.spark.api.java.JavaRDD;

 import org.apache.spark.api.java.JavaSparkContext;

 import org.apache.spark.api.java.function.Function;

 import org.apache.spark.mllib.classification.NaiveBayes;

 import org.apache.spark.mllib.classification.NaiveBayesModel;

 import org.apache.spark.mllib.regression.LabeledPoint;

 import org.apache.spark.mllib.util.MLUtils;

 import scala.Tuple2;

 /**

  * TODO

  *

  * @ClassName: example

  * @author: DingH

  * @since: 2019/4/10 10:04

  */

 public class example {

     public static void main(String[] args) {

         SparkConf conf = new SparkConf().setMaster("local").setAppName("naiveBayesExample");

         JavaSparkContext jsc = new JavaSparkContext(conf);

         String path = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_libsvm_data.txt";

         JavaRDD<LabeledPoint> data = MLUtils.loadLibSVMFile(jsc.sc(), path).toJavaRDD();

         JavaRDD<LabeledPoint>[] split = data.randomSplit(new double[]{0.6, 0.4}, 12345L);

         JavaRDD<LabeledPoint> train = split[0];

         JavaRDD<LabeledPoint> test = split[1];

         final NaiveBayesModel model = NaiveBayes.train(train.rdd(), 1.0, "multinomial");

         JavaRDD<Tuple2<Double, Double>> predictionAndLabel = test.map(new Function<LabeledPoint, Tuple2<Double, Double>>() {

             public Tuple2<Double, Double> call(LabeledPoint labeledPoint) throws Exception {

                 double predict = model.predict(labeledPoint.features());

                 return new Tuple2<Double, Double>(predict, labeledPoint.label());

             }

         });

         double accuracy = predictionAndLabel.filter(new Function<Tuple2<Double, Double>, Boolean>() {

             public Boolean call(Tuple2<Double, Double> doubleDoubleTuple2) throws Exception {

                 return doubleDoubleTuple2._1().equals(doubleDoubleTuple2._2());

             }

         }).count() / (double) test.count();

         System.out.println("acucuracy is : " + accuracy);

         NaiveBayesModel model1 = NaiveBayesModel.load(jsc.sc(), "");

     }

 }

naive Bayes

六。isotonic regression

 package ML.ClassificationAndRegression.IsotonicRegression;

 import org.apache.spark.SparkConf;

 import org.apache.spark.api.java.JavaDoubleRDD;

 import org.apache.spark.api.java.JavaRDD;

 import org.apache.spark.api.java.JavaSparkContext;

 import org.apache.spark.api.java.function.Function;

 import org.apache.spark.mllib.regression.IsotonicRegression;

 import org.apache.spark.mllib.regression.IsotonicRegressionModel;

 import scala.Tuple2;

 import scala.Tuple3;

 /**

  * TODO

  *

  * @ClassName: example

  * @author: DingH

  * @since: 2019/4/10 10:31

  */

 public class example {

     public static void main(String[] args) {

         SparkConf conf = new SparkConf().setMaster("local").setAppName("isotonicRegressionExample");

         JavaSparkContext jsc = new JavaSparkContext(conf);

         String path = "D:\\IdeaProjects\\SimpleApp\\src\\main\\resources\\data\\mllib\\sample_isotonic_regression_data.txt";

         JavaRDD<String> data = jsc.textFile(path);

         JavaRDD<Tuple3<Double, Double, Double>> parsedData = data.map(new Function<String, Tuple3<Double, Double, Double>>() {

             public Tuple3<Double, Double, Double> call(String s) throws Exception {

                 String[] strings = s.split(",");

                 return new Tuple3<Double, Double, Double>(Double.parseDouble(strings[0]), Double.parseDouble(strings[1]), 1.0);

             }

         });

         JavaRDD<Tuple3<Double, Double, Double>>[] split = parsedData.randomSplit(new double[]{0.7, 0.3}, 1234L);

         JavaRDD<Tuple3<Double, Double, Double>> train = split[0];

         JavaRDD<Tuple3<Double, Double, Double>> test = split[1];

         final IsotonicRegressionModel model = new IsotonicRegression().setIsotonic(true).run(train);

         JavaRDD<Tuple2<Double, Double>> preditionAndLabel = test.map(new Function<Tuple3<Double, Double, Double>, Tuple2<Double, Double>>() {

             public Tuple2<Double, Double> call(Tuple3<Double, Double, Double> doubleDoubleDoubleTuple3) throws Exception {

                 double predict = model.predict(doubleDoubleDoubleTuple3._1());

                 return new Tuple2<Double, Double>(predict, doubleDoubleDoubleTuple3._2());

             }

         });

         Double MSE = new JavaDoubleRDD(preditionAndLabel.map(new Function<Tuple2<Double, Double>, Object>() {

             public Object call(Tuple2<Double, Double> doubleDoubleTuple2) throws Exception {

                 return Math.pow(doubleDoubleTuple2._1() - doubleDoubleTuple2._2(), 2.0);

             }

         }).rdd()).mean();

         System.out.println("Mean Squared Error = " + MSE);

     }

 }

isotonic regression

spark MLlib Classification and regression 学习的更多相关文章

Spark MLlib Deep Learning Convolution Neural Network (深度学习-卷积神经网络)3.1
3.Spark MLlib Deep Learning Convolution Neural Network (深度学习-卷积神经网络)3.1 http://blog.csdn.net/sunbow0 ...
Spark MLlib Deep Learning Deep Belief Network (深度学习-深度信念网络)2.1
Spark MLlib Deep Learning Deep Belief Network (深度学习-深度信念网络)2.1 http://blog.csdn.net/sunbow0 Spark ML ...
Spark入门实战系列--8.Spark MLlib（上）--机器学习及SparkMLlib简介
[注]该系列文章以及使用到安装包/测试数据可以在<倾情大奉送--Spark入门实战系列>获取 .机器学习概念 1.1 机器学习的定义在维基百科上对机器学习提出以下几种定义: l“机器学 ...
Spark MLlib 机器学习
本章导读机器学习(machine learning, ML)是一门涉及概率论.统计学.逼近论.凸分析.算法复杂度理论等多领域的交叉学科.ML专注于研究计算机模拟或实现人类的学习行为,以获取新知识.新 ...
Spark MLlib架构解析（含分类算法、回归算法、聚类算法和协同过滤）
Spark MLlib架构解析 MLlib的底层基础解析 MLlib的算法库分析分类算法回归算法聚类算法协同过滤 MLlib的实用程序分析从架构图可以看出MLlib主要包含三个部分: 底层基 ...
spark Mllib基本功系列编程入门之 SVM实现分类
话不多说.直接上代码咯.欢迎交流. /** * Created by whuscalaman on 1/7/16. */import org.apache.spark.{SparkConf, Spar ...
spark Mllib SVM实例
Mllib SVM实例 1.数据数据格式为:标签, 特征1 特征2 特征3…… 0 128:51 129:159 130:253 131:159 132:50 155:48 156:238 157: ...
spark MLlib实现的基于朴素贝叶斯（NaiveBayes）的中文文本自动分类
1.自动文本分类是对大量的非结构化的文字信息(文本文档.网页等)按照给定的分类体系,根据文字信息内容分到指定的类别中去,是一种有指导的学习过程. 分类过程采用基于统计的方法和向量空间模型可以对常见的文 ...
在Java Web中使用Spark MLlib训练的模型
PMML是一种通用的配置文件,只要遵循标准的配置文件,就可以在Spark中训练机器学习模型,然后再web接口端去使用.目前应用最广的就是基于Jpmml来加载模型在javaweb中应用,这样就可以实现跨 ...

随机推荐

金融量化之Tushare模块
一.介绍 Tushare是一个免费.开源的python财经数据接口包.主要实现对股票等金融数据从数据采集.清洗加工到数据存储的过程,能够为金融分析人员提供快速.整洁.和多样的便于分析的数据,为他们 ...
Codeforces Global Round 2
A:答案一定包含首位或末位. #include<iostream> #include<cstdio> #include<cmath> #include<cst ...
docker-跨主机存储
容器分类从业务数据的角度看,容器可以分为两类:无状态(stateless)容器和有状态(stateful)容器. 无状态是指容器在运行过程中不需要保存数据,每次访问的结果不依赖上一次访问,比如提供静 ...
Python【第四篇】函数、内置函数、递归、装饰器、生成器和迭代器
一.函数函数是指将一组语句的集合通过一个名字(函数名)封装起来,要想执行这个函数,只需调用其函数名即可特性: 减少重复代码使程序变的可扩展使程序变得易维护 1.定义 def 函数名(参数): ...
shell 基础（二）变量
1. shell变量的定义 1)Shell 支持以下三种定义变量的方式: variable=value variable='value' variable="value" 特点 1 ...
【译】1. Java反射——引言
原文地址:http://tutorials.jenkov.com/java-reflection/index.html *By Jakob Jenkov Java的反射机制使得它可以在运行时检查类.接 ...
OS + macOS Mojave 10.14.4 / sushi / ssh-keygen / ssh-copy-id
s 系统版本: macOS 10.14.4 (18E226) 内核版本: Darwin 18.5.0 型号名称: Mac mini 2014 型号标识符: Macmini7,1 处理器名称: Inte ...
linux,pthread（转）
互斥量.条件变量与pthread_cond_wait()函数的使用,详解(二) 1.Linux“线程” 进程与线程之间是有区别的,不过linux内核只提供了轻量进程的支持,未实现线程模型.Linu ...
CentOS7部署Dotnet Core2.1
前言笔者在毫无Linux部署.net core的经验下,第一次用了15分钟完成部署,第二次在生产环境用了5分钟.下文将说明如何在CentOS7下完成.NetCore2.1的部署,包括如何创建ASP. ...
INI配置文件的格式
为什么要用INI文件?如果我们程序没有任何配置文件时,这样的程序对外是全封闭的,一旦程序需要修改一些参数必须要修改程序代码本身并重新编译,这样很不好,所以要用配置文件,让程序出厂后还能根据需要进行必要 ...

spark MLlib Classification and regression 学习

spark MLlib Classification and regression 学习的更多相关文章

随机推荐

热门专题