updated data, fixed bugs in script

HDI/Data/Loan_Data10000.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3c1a6af8d90609b03e5a3f5fd7b3da086244e1fc0c449ee4cebd918c16f731f
-size 81893495
+oid sha256:5972b70bfd8d8f9796a34020d8a16c7d200f5566339df2c220723f7a570a1003
+size 80441126

HDI/Data/Loan_Data100000.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f79aa81d2e77cfbfee2e0c0b150dfb6862dbbf5e30952e8c539b0baf97fe5314
-size 824443460
+oid sha256:5186585ba33a2f3138eb9fcff12c410e265796b197d3b04ffad7dad0f0e5704c
+size 808196811

HDI/RSparkCluster/loanchargeoff_deployment.R

@@ -137,7 +137,11 @@ result_string <- api_string$loan_web_scoring(
   userName = Sys.info()[["user"]],
   Stage = "Web")
 
-Loan_Data_df <- rxImport(RxTextData(file = file.path(HDFSDataDir, Loan_Data), fileSystem = RxHdfsFileSystem()))
+Loan_Data_df <- rxImport(RxTextData(file = file.path(HDFSDataDir, Loan_Data), 
+                                    missingValueString = "M", 
+                                    quotedDelimiters = TRUE,
+                                    fileSystem = RxHdfsFileSystem()))
+                                    
 # Verify the data frame input case
 result_string <- api_frame$loan_web_scoring(
   Loan_Data = Loan_Data_df,

HDI/RSparkCluster/loanchargeoff_web_scoring.R

@@ -87,7 +87,8 @@ HDFSDataDir <- "/LoanChargeOff/Data"
 HDFSWorkDir <- "/LoanChargeOff/web"
 
 # input data 
-Loan_Data_df <- read.csv("https://raw.githubusercontent.com/Microsoft/r-server-loan-chargeoff/master/HDI/Data/Loan_Data1000.csv")
+download.file("https://raw.githubusercontent.com/Microsoft/r-server-loan-chargeoff/master/HDI/Data/Loan_Data1000.csv", destfile = Loan_Data)
+Loan_Data_df <- read.csv(Loan_Data)
 
 # Call the web service
 result_string <- api_frame$loan_web_scoring(