LogisticRegression

The LogisticRegression algorithm uses ML APIs.

• Input/Output
  1. Package name: package org.apache.spark.ml.classification
  2. Class name: LogisticRegression
  3. Method name: fit
  4. Input: training sample data (Dataset[_]). Mandatory fields are as follows:

     Parameter	Type	Default Value	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:

     1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

     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.

     Parameter	Type	Default Value	Description
     predictionCol	Double	prediction	Predicted Label

• Sample usage

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

  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)