org.apache.spark.ml.r

Class RWrapperUtils

Object

org.apache.spark.ml.r.RWrapperUtils

public class RWrapperUtils
extends Object

Constructor Summary

Constructors

Constructor and Description
RWrapperUtils()

Method Summary

Modifier and Type	Method and Description
static void	checkDataColumns(RFormula rFormula, Dataset<?> data) DataFrame column check.
static scala.Tuple2<String[],String[]>	getFeaturesAndLabels(RFormulaModel rFormulaModel, Dataset<?> data) Get the feature names and original labels from the schema of DataFrame transformed by RFormulaModel.

Methods inherited from class Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

RWrapperUtils

public RWrapperUtils()

Method Detail

checkDataColumns

public static void checkDataColumns(RFormula rFormula,
                                    Dataset<?> data)

DataFrame column check. When loading libsvm data, default columns "features" and "label" will be added. And "features" would conflict with RFormula default feature column names. Here is to change the column name to avoid "column already exists" error.

Parameters:

rFormula - RFormula instance

data - Input dataset

getFeaturesAndLabels

public static scala.Tuple2<String[],String[]> getFeaturesAndLabels(RFormulaModel rFormulaModel,
                                                                   Dataset<?> data)

Get the feature names and original labels from the schema of DataFrame transformed by RFormulaModel.

Parameters:

rFormulaModel - The RFormulaModel instance.

data - Input dataset.

Returns:

The feature names and original labels.