Added helper functions for Renjin compatibility layer

build.sbt

@@ -28,11 +28,11 @@ val baseSettings = Seq(
 
 lazy val commonSettings = Seq(
   libraryDependencies ++= (
-    baseDependencies ++ apacheSpark ++
+    baseDependencies ++ apacheSparkDependency ++
       replDependency ++ loggingDependency ++
       linearAlgebraDependencies ++ chartsDependencies ++
       tinkerpopDependency ++ notebookInterfaceDependency ++
-      openMLDependency ++ rejinDependency)
+      openMLDependency ++ rejinDependency ++ rPackages)
 )
 
 lazy val pipes = (project in file("dynaml-pipes")).settings(baseSettings:_*)

conf/DynaMLInit.scala

@@ -1,21 +1,47 @@
-//External Imports
+/*
+* External Imports
+* */
+
+//Import breeze for linear algebra
 import breeze.linalg._
+//Apache Spark for big data support
 import org.apache.spark.SparkContext
 import org.apache.spark.SparkConf
+//Load Wisp-Highcharts for plotting
 import com.quantifind.charts.Highcharts._
+//Import spire implicits for definition of
+//fields, algebraic structures on primitive types
 import spire.implicits._
-//DynaML imports
+/*
+* DynaML imports
+* */
 import io.github.mandar2812.dynaml.analysis.VectorField
+//The pipes API
+import io.github.mandar2812.dynaml.pipes._
+import io.github.mandar2812.dynaml.DynaMLPipe
+import io.github.mandar2812.dynaml.DynaMLPipe._
+//Load the DynaML model api members
 import io.github.mandar2812.dynaml.models._
 import io.github.mandar2812.dynaml.models.neuralnets._
 import io.github.mandar2812.dynaml.models.svm._
 import io.github.mandar2812.dynaml.models.lm._
+//Utility functions
 import io.github.mandar2812.dynaml.utils
+//Kernels for GP,SVM models
 import io.github.mandar2812.dynaml.kernels._
+//Shell examples
 import io.github.mandar2812.dynaml.examples._
-import io.github.mandar2812.dynaml.pipes._
-import io.github.mandar2812.dynaml.DynaMLPipe
-import io.github.mandar2812.dynaml.DynaMLPipe._
+//Load neural net primitives
 import io.github.mandar2812.dynaml.models.neuralnets.TransferFunctions._
+//The probability API
 import io.github.mandar2812.dynaml.probability._
+//OpenML support
 import io.github.mandar2812.dynaml.openml.OpenML
+//Renjin imports
+import javax.script._
+import org.renjin.script._
+import org.renjin.sexp._
+val r_engine_factory = new RenjinScriptEngineFactory()
+implicit val renjin = r_engine_factory.getScriptEngine()
+val r: String => SEXP = (s: String) => renjin.eval(s).asInstanceOf[SEXP]
+val R: java.io.File => SEXP = (f: java.io.File) => renjin.eval(f).asInstanceOf[SEXP]

dynaml-core/src/main/scala-2.11/io/github/mandar2812/dynaml/DynaMLPipe.scala

@@ -25,8 +25,10 @@ import io.github.mandar2812.dynaml.models.ParameterizedLearner
 import io.github.mandar2812.dynaml.models.gp.AbstractGPRegressionModel
 import io.github.mandar2812.dynaml.optimization.{CoupledSimulatedAnnealing, GPMLOptimizer, GloballyOptWithGrad, GridSearch}
 import io.github.mandar2812.dynaml.pipes.{DataPipe, ReversibleScaler, Scaler, StreamDataPipe}
-import io.github.mandar2812.dynaml.utils.{GaussianScaler, MinMaxScaler, MVGaussianScaler}
+import io.github.mandar2812.dynaml.utils.{GaussianScaler, MVGaussianScaler, MinMaxScaler}
 import org.apache.log4j.Logger
+import org.renjin.script.RenjinScriptEngine
+import org.renjin.sexp._
 
 /**
   * @author mandar2812 datum 3/2/16.
@@ -52,6 +54,16 @@ object DynaMLPipe {
     * */
   val fileToStream = DataPipe(utils.textFileToStream _)
 
+  def csvToRDF(df: String, sep: Char)(implicit renjin: RenjinScriptEngine): DataPipe[String, ListVector] =
+    DataPipe((file: String) => renjin
+      .eval(df+""" <- read.csv(""""+file+"""", sep = '"""+sep+"""')""")
+      .asInstanceOf[ListVector])
+
+  def rdfToGLM(modelName: String, y: String, xs: Array[String])(implicit renjin: RenjinScriptEngine)
+  : DataPipe[String, ListVector] = DataPipe((df: String) => renjin
+    .eval(modelName+" <- lm("+y+" ~ "+xs.mkString(" + ")+", "+df+")")
+    .asInstanceOf[ListVector])
+
   /**
     * Writes a [[Stream]] of [[String]] to
     * a file.

project/Dependencies.scala

@@ -22,7 +22,7 @@ object Dependencies {
     "com.github.scopt" % "scopt_2.11" % "3.5.0"
   )
 
-  val apacheSpark = Seq(
+  val apacheSparkDependency = Seq(
     "javax.servlet" % "javax.servlet-api" % "3.1.0" % "test",
     "org.apache.spark" % "spark-core_2.11" % "2.0.0" % "compile",
     "org.apache.spark" % "spark-mllib_2.11" % "2.0.0" % "compile"
@@ -56,5 +56,10 @@ object Dependencies {
     "org.renjin" % "renjin-script-engine" % "0.8.2297"
   )
 
+  val rPackages = Seq(
+    "org.renjin.cran" % "plyr" % "1.8.3-renjin-10",
+    "org.renjin.cran" % "abc" % "2.1-b274"
+  )
+
   val notebookInterfaceDependency = Seq()
 }