Spark Mllib决策树示例三

需要数据源可以留言

object DecisionTreeClassification {
  def main(args: Array[String]): Unit = {
    System.setProperty("hadoop.home.dir","E:\tools\hadoop\hadoop-common-2.6.0-bin-master\hadoop-common-2.6.0-bin-master")
    val spark = SparkSession.builder().master("local").appName("Naive_bayes").getOrCreate()

    val sc = spark.sparkContext

    // Load and parse the data file.
    val data = MLUtils.loadLibSVMFile(sc, "data/汽车数据样本.txt")
    // Split the data into training and test sets (30% held out for testing)
    val splits: Array[RDD[LabeledPoint]] = data.randomSplit(Array(0.7, 0.3))
    val (trainingData, testData) = (splits(0), splits(1))

    // Train a DecisionTree model.
    //  Empty categoricalFeaturesInfo indicates all features are continuous.
    //指定分类的类别
    val numClasses = 2
    //设置离散变量，没有指定的话就是连续数据。如果不是连续数据说明数据已经离散化了，可以分类了。
    //这个参数就是告诉决策树，我们这边指定的数据，都不是连续数据，换言之，没有指定的数据就是连续数据。
    //不是连续数据就说明已经离散化了，是可以分类的。0号特征有四个类别，1号特征也有四个类别。
    val categoricalFeaturesInfo = Map[Int, Int](0->4,1->4,2->3,3->3)
    val impurity = "entropy"//gini & entropy两种选择，一个是熵一个是基尼系数
    //预剪枝，最大深度。防止模型过拟合。
    val maxDepth = 5
    val maxBins = 32//离散化程度处理连续数据，相当于%32，这个值减少也等同于剪枝

    val model = DecisionTree.trainClassifier(trainingData, numClasses, categoricalFeaturesInfo,
      impurity, maxDepth, maxBins)

    // evaluate model on test instances and compute test error
    val labelAndPreds: RDD[(Double, Double)] = testData.map { point =>
      val prediction = model.predict(point.features)
      (point.label, prediction)
    }
    val testErr = labelAndPreds.filter(r => r._1 != r._2).count().toDouble / testData.count()
    println(s"Test Error = $testErr")
    //决策树说白了就是一串if else
    println(s"Learned classification tree model:n ${model.toDebugString}")

    // Save and load model
    model.save(sc, "target/tmp/myDecisionTreeClassificationModel")
//    val sameModel = DecisionTreeModel.load(sc, "target/tmp/myDecisionTreeClassificationModel")


    spark.stop()
  }