LogisticRegression

ML classification API

• Function

  This type of APIs is used to import sample data in dataset format, call the fit API, and output the LogisticRegression model.

• Input and output
  1. Package name: package org.apache.spark.ml.classification
  2. Class name: LogisticRegression
  3. Method name: fit
  4. Input: training sample data (Dataset[_]). The following are mandatory fields.

     Param name	Type(s)	Default	Description
     labelCol	Double	"label"	Label, require: 1) label == label.toInt 2) label >= 0
     featuresCol	Vector	"features"	Feature label

  5. Algorithm parameters

     Algorithm Parameter

     def setRegParam(value: Double): LogisticRegression.this.type def setElasticNetParam(value: Double): LogisticRegression.this.type def setMaxIter(value: Int): LogisticRegression.this.type def setTol(value: Double): LogisticRegression.this.type def setFitIntercept(value: Boolean): LogisticRegression.this.type def setFamily(value: String): LogisticRegression.this.type def setStandardization(value: Boolean): LogisticRegression.this.type override def setThreshold(value: Double): LogisticRegression.this.type def setWeightCol(value: String): LogisticRegression.this.type override def setThresholds(value: Array[Double]): LogisticRegression.this.type def setAggregationDepth(value: Int): LogisticRegression.this.type def setLowerBoundsOnCoefficients(value: Matrix): LogisticRegression.this.type def setUpperBoundsOnCoefficients(value: Matrix): LogisticRegression.this.type def setLowerBoundsOnIntercepts(value: Vector): LogisticRegression.this.type def setUpperBoundsOnIntercepts(value: Vector):LogisticRegression.this.type

     

     An example is provided as follows:

     import org.apache.spark.ml.param.{ParamMap, ParamPair}
     
     val logR = new LogisticRegression()
     // Define the def fit(dataset: Dataset[_], paramMap: ParamMap) API parameter.
     val paramMap = ParamMap(logR.maxIter -> maxIter)
     .put(logR.regParam, regParam)
     
     // Define the def fit(dataset: Dataset[_], paramMaps: Array[ParamMap]): API parameter.
     val paramMaps: Array[ParamMap] = new Array[ParamMap](2)
     for (i <- 0 to  2) {
     paramMaps(i) = ParamMap(logR.maxIter -> maxIter)
     .put(logR.regParam, regParam)
     }// Assign a value to paramMaps.
     
     // Define the def fit(dataset: Dataset[_], firstParamPair: ParamPair[_], otherParamPairs: ParamPair[_]*) API parameter.
     val regParamPair = ParamPair(logR.regParam, regParam)
     val maxIterParamPair = ParamPair(logR.maxIter, maxIter)
     val tolParamPair = ParamPair(logR.tol, tol)
     
     // Call the fit APIs.
     model = logR.fit(trainingData)
     model = logR.fit(trainingData, paramMap)
     models = logR.fit(trainingData, paramMaps)
     model = logR.fit(trainingData, regParamPair, maxIterParamPair, tolParamPair)

  6. Output: LogisticRegressionModel. The output in model prediction is as follows.

     Param name	Type(s)	Default	Description
     predictionCol	Double	"prediction"	Predicted Label

• Sample usage

  import org.apache.spark.ml.classification.LogisticRegression
  
  // Load training data
  val training = spark.read.format("libsvm").load("data/mllib/sample_libsvm_data.txt")
  
  val lr = new LogisticRegression()
  .setMaxIter(10)
  .setRegParam(0.3)
  .setElasticNetParam(0.8)
  
  // Fit the model
  val lrModel = lr.fit(training)
  
  // Print the coefficients and intercept for logistic regression
  println(s"Coefficients: ${lrModel.coefficients} Intercept: ${lrModel.intercept}")
  
  // We can also use the multinomial family for binary classification
  val mlr = new LogisticRegression()
  .setMaxIter(10)
  .setRegParam(0.3)
  .setElasticNetParam(0.8)
  .setFamily("multinomial")
  
  val mlrModel = mlr.fit(training)