ramnov

##########################################################
#       Create & Test a Logistic Regression Model        #
##########################################################

# load mrsdeploy package on R Server     
library(mrsdeploy)

# Use logistic regression equation of vehicle transmission 
# in the data set mtcars to estimate the probability of 
# a vehicle being fitted with a manual transmission 

ramnov

# Generate Service Consumption Client Library in CSharp from swagger.json
install.packages("httr", dependencies = TRUE)
library(httr)
serviceName <- "ManualTransmissionService"
swaggerFileLocation <- "C:/Users/azureuser/Documents/swagger.json"
swagger <- readChar(swaggerFileLocation, file.info(swaggerFileLocation)$size)
r <- httr::POST("http://generator.swagger.io/api/gen/clients/csharp",
                add_headers("Content-Type" = "application/json"),
                body = paste("{\"options\": { \"packageName\" : \"", serviceName, "\" } ,",
                             "\"spec\":", swagger, "}"))

ramnov

# Estimate a One-Class SVM model
trainRows <- c(1:30, 51:80, 101:130)
testRows = !(1:150 %in% trainRows)
trainIris <- iris[trainRows,]
testIris <- iris[testRows,]

trainIris$id <- seq.int(nrow(trainIris))
svmModel <- rxOneClassSvm(
  formula = ~Sepal.Length + Sepal.Width + Petal.Length + Petal.Width,
  data = trainIris)

ramnov

using System;
using System.Collections.Generic;
using System.Linq;
using System.Security.Cryptography.X509Certificates;

namespace InstallCert
{
    public class Program
    {
        public static void Main(string[] args)

ramnov

library(httr)
r <- httr::GET("https://<yourstorename>.azuredatalakestore.net/webhdfs/v1/mytempdir/iris.csv?op=OPEN&read=true",
                add_headers(Authorization = "Bearer <AD AUTH TOKEN>"),
                write_disk("iris.csv", overwrite=TRUE), progress())
irisData <- rxImport("iris.csv")

ramnov

# Work with HDFS data in local compute context
rxHadoopMakeDir("/share/SampleData")
rxHadoopCopyFromLocal(file.path(dataPath = rxGetOption("sampleDataDir"), "AirlineDemoSmall.csv"), "/share/SampleData")
airDS <- RxTextData(file = "/share/SampleData/AirlineDemoSmall.csv", missingValueString = "M", 
                    fileSystem = RxHdfsFileSystem())
rxSetComputeContext("local")                    
adsSummary <- rxSummary(~ArrDelay+CRSDepTime+DayOfWeek, data = airDS)
print(adsSummary)

ramnov

# Spark Compute Context
sparkCC <- rxSparkConnect()
rxHadoopMakeDir("/share/SampleData")
rxHadoopCopyFromLocal(file.path(dataPath = rxGetOption("sampleDataDir"), "AirlineDemoSmall.csv"), "/share/SampleData")
airDS <- RxTextData(file = "/share/SampleData/AirlineDemoSmall.csv", missingValueString = "M", 
                    fileSystem = RxHdfsFileSystem())
adsSummary <- rxSummary(~ArrDelay+CRSDepTime+DayOfWeek, data = airDS)
print(adsSummary)
rxSparkDisconnect(sparkCC)

ramnov

# Hadoop Compute Context
rxSetComputeContext(RxHadoopMR())
rxHadoopMakeDir("/share/SampleData")
rxHadoopCopyFromLocal(file.path(dataPath = rxGetOption("sampleDataDir"), "AirlineDemoSmall.csv"), "/share/SampleData")
airDS <- RxTextData(file = "/share/SampleData/AirlineDemoSmall.csv", missingValueString = "M", 
                    fileSystem = RxHdfsFileSystem())
adsSummary <- rxSummary(~ArrDelay+CRSDepTime+DayOfWeek, data = airDS)
print(adsSummary)

ramnov

# Local Compute Context
rxSetComputeContext("local")
airDS <- RxTextData(file.path(rxGetOption("sampleDataDir"), "AirlineDemoSmall.csv"))
adsSummary <- rxSummary(~ArrDelay+CRSDepTime+DayOfWeek, data = airDS)
print(adsSummary)

ramnov

# Set Seed
set.seed(12345)

# AirlineDemoSmall XDF File
airline <- RxXdfData(file.path(rxGetOption("sampleDataDir"), "AirlineDemoSmall.xdf"))

# Define a split function which splits the dataset into 75% train and 25% test using rxSplit
".split" <- function(keys, data)
{
  dataSplit <- rxSplit(data,