Spark 多项式逻辑回归__多分类

Posted soyosuyang

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了Spark 多项式逻辑回归__多分类相关的知识,希望对你有一定的参考价值。

package Spark_MLlib

import org.apache.spark.ml.Pipeline
import org.apache.spark.ml.classification.{BinaryLogisticRegressionSummary, LogisticRegression, LogisticRegressionModel}
import org.apache.spark.ml.evaluation.MulticlassClassificationEvaluator
import org.apache.spark.ml.feature.{IndexToString, StringIndexer, VectorIndexer}
import org.apache.spark.ml.linalg.Vectors
import org.apache.spark.sql.SparkSession


object 多项式逻辑回归__多分类 {
  val spark=SparkSession.builder().master("local").getOrCreate()
  import spark.implicits._  //支持把一个RDD隐式转换为一个DataFrame
  def main(args: Array[String]): Unit = {
    val df =spark.sparkContext.textFile("file:///home/soyo/桌面/spark编程测试数据/soyo.txt")
      .map(_.split(",")).map(x=>data_schema(Vectors.dense(x(0).toDouble,x(1).toDouble,x(2).toDouble,x(3).toDouble),x(4))).toDF()
   // df.show(150)
    val labelIndexer=new StringIndexer().setInputCol("label").setOutputCol("indexedLabel").fit(df)
    val featureIndexer=new VectorIndexer().setInputCol("features").setOutputCol("indexedFeatures").fit(df)  //目的在特征向量中建类别索引
    val Array(trainData,testData)=df.randomSplit(Array(0.7,0.3))
    val lr=new LogisticRegression().setLabelCol("indexedLabel").setFeaturesCol("indexedFeatures").setMaxIter(10).setRegParam(0.3).setElasticNetParam(0.8).setFamily("multinomial")//设置elasticnet混合参数为0.8,setFamily("multinomial"):设置为多项逻辑回归,不设置setFamily为二项逻辑回归
    val labelConverter=new IndexToString().setInputCol("prediction").setOutputCol("predictionLabel").setLabels(labelIndexer.labels)

    val lrPipeline=new Pipeline().setStages(Array(labelIndexer,featureIndexer,lr,labelConverter))
    val lrPipeline_Model=lrPipeline.fit(trainData)
    val lrPrediction=lrPipeline_Model.transform(testData)
    lrPrediction.show(150)
    // lrPrediction.take(100).foreach(println)
    //模型评估
    val evaluator=new MulticlassClassificationEvaluator().setLabelCol("indexedLabel").setPredictionCol("prediction")
    val lrAccuracy=evaluator.evaluate(lrPrediction)
    println("准确率为: "+lrAccuracy)
    val lrError=1-lrAccuracy
    println("错误率为: "+lrError)
    val LRmodel=lrPipeline_Model.stages(2).asInstanceOf[LogisticRegressionModel]
    println("二项逻辑回归模型系数矩阵: "+LRmodel.coefficientMatrix)
    println("二项逻辑回归模型的截距向量: "+LRmodel.interceptVector)
    println("类的数量(标签可以使用的值): "+LRmodel.numClasses)
    println("模型所接受的特征的数量: "+LRmodel.numFeatures)
  //多项式逻辑回归不包含对模型的摘要总结
    println(LRmodel.hasSummary)

  }

}

结果:

+-----------------+-----+------------+-----------------+--------------------+--------------------+----------+---------------+
|         features|label|indexedLabel|  indexedFeatures|       rawPrediction|         probability|prediction|predictionLabel|
+-----------------+-----+------------+-----------------+--------------------+--------------------+----------+---------------+
|[4.4,3.2,1.3,0.2]|soyo1|         1.0|[4.4,3.2,1.3,0.2]|[0.06313829278191...|[0.23858281707128...|       1.0|          soyo1|
|[4.6,3.4,1.4,0.3]|soyo1|         1.0|[4.6,3.4,1.4,0.3]|[0.06313829278191...|[0.23750012598226...|       1.0|          soyo1|
|[4.7,3.2,1.6,0.2]|soyo1|         1.0|[4.7,3.2,1.6,0.2]|[0.06313829278191...|[0.24710416166321...|       1.0|          soyo1|
|[4.8,3.4,1.6,0.2]|soyo1|         1.0|[4.8,3.4,1.6,0.2]|[0.06313829278191...|[0.23716995683018...|       1.0|          soyo1|
|[4.8,3.4,1.9,0.2]|soyo1|         1.0|[4.8,3.4,1.9,0.2]|[0.06313829278191...|[0.24567798276462...|       1.0|          soyo1|
|[4.9,2.4,3.3,1.0]|soyo2|         0.0|[4.9,2.4,3.3,1.0]|[0.06313829278191...|[0.38071131817453...|       0.0|          soyo2|
|[5.0,3.2,1.2,0.2]|soyo1|         1.0|[5.0,3.2,1.2,0.2]|[0.06313829278191...|[0.23576075216827...|       1.0|          soyo1|
|[5.0,3.5,1.3,0.3]|soyo1|         1.0|[5.0,3.5,1.3,0.3]|[0.06313829278191...|[0.22978111243935...|       1.0|          soyo1|
|[5.2,4.1,1.5,0.1]|soyo1|         1.0|[5.2,4.1,1.5,0.1]|[0.06313829278191...|[0.19523110424215...|       1.0|          soyo1|
|[5.4,3.9,1.3,0.4]|soyo1|         1.0|[5.4,3.9,1.3,0.4]|[0.06313829278191...|[0.21630436073381...|       1.0|          soyo1|
|[5.5,2.4,3.8,1.1]|soyo2|         0.0|[5.5,2.4,3.8,1.1]|[0.06313829278191...|[0.39807479409636...|       0.0|          soyo2|
|[5.5,2.5,4.0,1.3]|soyo2|         0.0|[5.5,2.5,4.0,1.3]|[0.06313829278191...|[0.40810357240132...|       0.0|          soyo2|
|[5.6,2.8,4.9,2.0]|soyo3|         2.0|[5.6,2.8,4.9,2.0]|[0.06313829278191...|[0.44454733071968...|       0.0|          soyo2|
|[5.7,2.9,4.2,1.3]|soyo2|         0.0|[5.7,2.9,4.2,1.3]|[0.06313829278191...|[0.39634982244233...|       0.0|          soyo2|
|[5.8,2.6,4.0,1.2]|soyo2|         0.0|[5.8,2.6,4.0,1.2]|[0.06313829278191...|[0.39930520027794...|       0.0|          soyo2|
|[5.8,2.7,4.1,1.0]|soyo2|         0.0|[5.8,2.7,4.1,1.0]|[0.06313829278191...|[0.38762610877473...|       0.0|          soyo2|
|[5.8,2.7,5.1,1.9]|soyo3|         2.0|[5.8,2.7,5.1,1.9]|[0.06313829278191...|[0.44792417666537...|       0.0|          soyo2|
|[5.9,3.0,5.1,1.8]|soyo3|         2.0|[5.9,3.0,5.1,1.8]|[0.06313829278191...|[0.43418725338764...|       0.0|          soyo2|
|[6.0,2.2,4.0,1.0]|soyo2|         0.0|[6.0,2.2,4.0,1.0]|[0.06313829278191...|[0.40634099537710...|       0.0|          soyo2|
|[6.0,2.7,5.1,1.6]|soyo2|         0.0|[6.0,2.7,5.1,1.6]|[0.06313829278191...|[0.43688076686419...|       0.0|          soyo2|
|[6.0,3.4,4.5,1.6]|soyo2|         0.0|[6.0,3.4,4.5,1.6]|[0.06313829278191...|[0.39704954911011...|       0.0|          soyo2|
|[6.2,2.2,4.5,1.5]|soyo2|         0.0|[6.2,2.2,4.5,1.5]|[0.06313829278191...|[0.43847273913421...|       0.0|          soyo2|
|[6.2,2.8,4.8,1.8]|soyo3|         2.0|[6.2,2.8,4.8,1.8]|[0.06313829278191...|[0.43518321759857...|       0.0|          soyo2|
|[6.3,2.7,4.9,1.8]|soyo3|         2.0|[6.3,2.7,4.9,1.8]|[0.06313829278191...|[0.44055947195014...|       0.0|          soyo2|
|[6.3,2.9,5.6,1.8]|soyo3|         2.0|[6.3,2.9,5.6,1.8]|[0.06313829278191...|[0.44715759200377...|       0.0|          soyo2|
|[6.3,3.4,5.6,2.4]|soyo3|         2.0|[6.3,3.4,5.6,2.4]|[0.06313829278191...|[0.45196576310313...|       0.0|          soyo2|
|[6.4,2.8,5.6,2.1]|soyo3|         2.0|[6.4,2.8,5.6,2.1]|[0.06313829278191...|[0.46017875340546...|       0.0|          soyo2|
|[6.4,2.8,5.6,2.2]|soyo3|         2.0|[6.4,2.8,5.6,2.2]|[0.06313829278191...|[0.46321910727428...|       0.0|          soyo2|
|[6.4,3.1,5.5,1.8]|soyo3|         2.0|[6.4,3.1,5.5,1.8]|[0.06313829278191...|[0.43862320280893...|       0.0|          soyo2|
|[6.4,3.2,4.5,1.5]|soyo2|         0.0|[6.4,3.2,4.5,1.5]|[0.06313829278191...|[0.40056786531830...|       0.0|          soyo2|
|[6.5,3.0,5.5,1.8]|soyo3|         2.0|[6.5,3.0,5.5,1.8]|[0.06313829278191...|[0.44199581778961...|       0.0|          soyo2|
|[6.6,2.9,4.6,1.3]|soyo2|         0.0|[6.6,2.9,4.6,1.3]|[0.06313829278191...|[0.40579282648595...|       0.0|          soyo2|
|[6.7,2.5,5.8,1.8]|soyo3|         2.0|[6.7,2.5,5.8,1.8]|[0.06313829278191...|[0.46287803722998...|       0.0|          soyo2|
|[6.7,3.0,5.2,2.3]|soyo3|         2.0|[6.7,3.0,5.2,2.3]|[0.06313829278191...|[0.45387841693477...|       0.0|          soyo2|
|[6.7,3.1,4.7,1.5]|soyo2|         0.0|[6.7,3.1,4.7,1.5]|[0.06313829278191...|[0.40924150360290...|       0.0|          soyo2|
|[6.7,3.3,5.7,2.5]|soyo3|         2.0|[6.7,3.3,5.7,2.5]|[0.06313829278191...|[0.45972648058424...|       0.0|          soyo2|
|[6.8,3.0,5.5,2.1]|soyo3|         2.0|[6.8,3.0,5.5,2.1]|[0.06313829278191...|[0.45251276088924...|       0.0|          soyo2|
|[6.8,3.2,5.9,2.3]|soyo3|         2.0|[6.8,3.2,5.9,2.3]|[0.06313829278191...|[0.45975331380088...|       0.0|          soyo2|
|[6.9,3.2,5.7,2.3]|soyo3|         2.0|[6.9,3.2,5.7,2.3]|[0.06313829278191...|[0.45642868507279...|       0.0|          soyo2|
|[7.2,3.0,5.8,1.6]|soyo3|         2.0|[7.2,3.0,5.8,1.6]|[0.06313829278191...|[0.44031726493318...|       0.0|          soyo2|
|[7.2,3.2,6.0,1.8]|soyo3|         2.0|[7.2,3.2,6.0,1.8]|[0.06313829278191...|[0.44483171938259...|       0.0|          soyo2|
|[7.6,3.0,6.6,2.1]|soyo3|         2.0|[7.6,3.0,6.6,2.1]|[0.06313829278191...|[0.47047723863543...|       0.0|          soyo2|
|[7.7,3.0,6.1,2.3]|soyo3|         2.0|[7.7,3.0,6.1,2.3]|[0.06313829278191...|[0.46845272424381...|       0.0|          soyo2|
|[7.7,3.8,6.7,2.2]|soyo3|         2.0|[7.7,3.8,6.7,2.2]|[0.06313829278191...|[0.45233124776236...|       0.0|          soyo2|
+-----------------+-----+------------+-----------------+--------------------+--------------------+----------+---------------+

准确率为: 0.36458333333333337
错误率为: 0.6354166666666666
二项逻辑回归模型系数矩阵: 3 x 4 CSCMatrix
(1,1) 0.35559564188466614
(1,2) -0.203185158868005
(1,3) -0.43876460704959996
(2,3) 0.0283914830858408
二项逻辑回归模型的截距向量: [0.06313829278191783,0.1708622138778958,-0.23400050665981365]
类的数量(标签可以使用的值): 3
模型所接受的特征的数量: 4
false

以上是关于Spark 多项式逻辑回归__多分类的主要内容,如果未能解决你的问题,请参考以下文章

使用 scikit learn 训练逻辑回归进行多类分类

如何使用多项逻辑回归解决多标签分类问题?

逻辑回归-4.添加多项式特征

如何使用多项逻辑回归模型来预测未来的观察结果

[机器学习与scikit-learn-21]:算法-逻辑回归-多项式非线性回归PolynomialFeatures与代码实现

Spark 多项式 Logistic 回归中的意外系数