Spark MLlib 貝葉斯分類演算法例項具體程式碼及執行過程詳解
阿新 • • 發佈：2018-12-19
import org.apache.log4j.{Level, Logger}
import org.apache.spark.{SparkConf, SparkContext}
import org.apache.spark.mllib.classification.{NaiveBayes, NaiveBayesModel}
import org.apache.spark.mllib.linalg.Vectors
import org.apache.spark.mllib.regression.LabeledPoint
//資料格式：類別，特徵1 特徵2 特徵3
//0,1 0 1
//1,0 2 0
object tmp_naive_bayes {
def main(args: Array[String]){
//1、構建spark物件
val conf = new SparkConf().setAppName("naive_bayes").setMaster("local")
val sc = new SparkContext(conf)
Logger.getRootLogger.setLevel(Level.WARN)
//2、讀取資料樣本
val data = sc.textFile("C://Users/wpguoc/Desktop/Spark MLlib/navie_bayes_data.txt")
val parsedData = data.map{ line =>
val parts = line.split(',')
LabeledPoint(parts(0).toDouble, Vectors.dense(parts(1).split(' ').map(_.toDouble)))
}
//3、樣本資料劃分訓練樣本和測試樣本
val splits = parsedData.randomSplit(Array(0.6, 0.4), seed = 11L)
val tran = splits(0)
val test = splits(1)
//4、新建貝葉斯分類模型，並訓練
val model = NaiveBayes.train(tran, lambda = 1.0, modelType = "multinomial")
//5、對測試樣本進行測試
val predictionAndLabel = test.map(p => (model.predict(p.features), p.label))
val print_predict = predictionAndLabel.take(20)
println("貝葉斯分類結果：" + "\n" + "prediction" + "\t" + "label")
for(i <- 0 to print_predict.length - 1){
println(print_predict(i)._1+"\t\t\t"+print_predict(i)._2)
}
val accuracy = 1.0*predictionAndLabel.filter(x =>x._1 == x._2).count() / test.count()
println("貝葉斯分類精度" + "\n" + "accuracy: Double = " + accuracy)
//6、儲存模型
val ModelPath = "C://Users/wpguoc/Desktop/Spark_MLlib/"
model.save(sc, ModelPath)
val sameModel = NaiveBayesModel.load(sc, ModelPath)
}
}
執行過程及結果
"C:\Program Files\Java\jdk1.8.0_181\bin\java" "-javaagent:D:\Software\IntelliJ IDEA 2017.2.2\lib\idea_rt.jar=63402:D:\Software\IntelliJ IDEA 2017.2.2\bin" -Dfile.encoding=UTF-8 -classpath "C:\Program Files\Java\jdk1.8.0_181\jre\lib\charsets.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\deploy.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\access-bridge-64.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\cldrdata.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\dnsns.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\jaccess.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\jfxrt.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\localedata.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\nashorn.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\sunec.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\sunjce_provider.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\sunmscapi.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\sunpkcs11.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\ext\zipfs.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\javaws.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\jce.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\jfr.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\jfxswt.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\jsse.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\management-agent.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\plugin.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\resources.jar;C:\Program Files\Java\jdk1.8.0_181\jre\lib\rt.jar;E:\Scala_workplace\scala菜鳥教程\production\Spark_Mllibe;C:\Program Files (x86)\scala\lib\scala-actors-migration.jar;C:\Program Files (x86)\scala\lib\scala-actors.jar;C:\Program Files (x86)\scala\lib\scala-library.jar;C:\Program Files (x86)\scala\lib\scala-reflect.jar;C:\Program Files (x86)\scala\lib\scala-swing.jar;D:\spark-1.6.3-bin-2.6.0-cdh5.7.0\lib\spark-assembly-1.6.3-hadoop2.6.0-cdh5.7.0.jar" tmp_naive_bayes Using Spark's default log4j profile: org/apache/spark/log4j-defaults.properties 18/12/19 17:32:04 INFO SparkContext: Running Spark version 1.6.3 18/12/19 17:32:04 WARN NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable 18/12/19 17:32:05 INFO SecurityManager: Changing view acls to: wpguoc 18/12/19 17:32:05 INFO SecurityManager: Changing modify acls to: wpguoc 18/12/19 17:32:05 INFO SecurityManager: SecurityManager: authentication disabled; ui acls disabled; users with view permissions: Set(wpguoc); users with modify permissions: Set(wpguoc) 18/12/19 17:32:05 INFO Utils: Successfully started service 'sparkDriver' on port 63424. 18/12/19 17:32:06 INFO Slf4jLogger: Slf4jLogger started 18/12/19 17:32:06 INFO Remoting: Starting remoting 18/12/19 17:32:06 INFO Remoting: Remoting started; listening on addresses :[akka.tcp://[email protected]:63437] 18/12/19 17:32:06 INFO Utils: Successfully started service 'sparkDriverActorSystem' on port 63437. 18/12/19 17:32:06 INFO SparkEnv: Registering MapOutputTracker 18/12/19 17:32:06 INFO SparkEnv: Registering BlockManagerMaster 18/12/19 17:32:06 INFO DiskBlockManager: Created local directory at C:\Users\wpguoc\AppData\Local\Temp\blockmgr-4d798e34-90b0-4ee9-a811-586a893f4818 18/12/19 17:32:06 INFO MemoryStore: MemoryStore started with capacity 1127.3 MB 18/12/19 17:32:06 INFO SparkEnv: Registering OutputCommitCoordinator 18/12/19 17:32:06 INFO Utils: Successfully started service 'SparkUI' on port 4040. 18/12/19 17:32:06 INFO SparkUI: Started SparkUI at http://192.168.66.80:4040 18/12/19 17:32:06 INFO Executor: Starting executor ID driver on host localhost 18/12/19 17:32:06 INFO Utils: Successfully started service 'org.apache.spark.network.netty.NettyBlockTransferService' on port 63444. 18/12/19 17:32:06 INFO NettyBlockTransferService: Server created on 63444 18/12/19 17:32:06 INFO BlockManagerMaster: Trying to register BlockManager 18/12/19 17:32:06 INFO BlockManagerMasterEndpoint: Registering block manager localhost:63444 with 1127.3 MB RAM, BlockManagerId(driver, localhost, 63444) 18/12/19 17:32:06 INFO BlockManagerMaster: Registered BlockManager 18/12/19 17:32:10 WARN BLAS: Failed to load implementation from: com.github.fommil.netlib.NativeSystemBLAS 18/12/19 17:32:10 WARN BLAS: Failed to load implementation from: com.github.fommil.netlib.NativeRefBLAS 貝葉斯分類結果： prediction label 0.0 0.0 0.0 0.0 2.0 2.0 2.0 2.0 2.0 2.0 貝葉斯分類精度 accuracy: Double = 1.0 SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder". SLF4J: Defaulting to no-operation (NOP) logger implementation SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details. 18/12/19 17:32:14 WARN ParquetRecordReader: Can not initialize counter due to context is not a instance of TaskInputOutputContext, but is org.apache.hadoop.mapreduce.task.TaskAttemptContextImpl Process finished with exit code 0