Team 1 Pull request

.Rproj.user/E6F84AE4/sources/prop/5446D37A

@@ -0,0 +1,4 @@
+{
+    "cursorPosition" : "45,0",
+    "scrollLine" : "28"
+}

.Rproj.user/E6F84AE4/sources/prop/INDEX

@@ -0,0 +1,6 @@
+~%2Fbmarketinggg%2FTeam1%2FR%2FPlot.R="85A4D0EF"
+~%2Fbmarketinggg%2FTeam1%2FR%2Fmodel.R="78826CD7"
+~%2Fbmarketinggg%2Fbmarketing.R="5446D37A"
+~%2Fbmarketinggggg%2FTeam1%2FR%2Fmodel.R="2432FCD0"
+~%2Fbmarketinggggg%2FTeam1%2FR%2Fpredict.R="4BB9E5C6"
+~%2Fbmarketinggggg%2Fbmarketing.R="A971490D"

.Rproj.user/E6F84AE4/sources/s-5EE927D0/B6001EF7

@@ -0,0 +1,25 @@
+{
+    "collab_server" : "",
+    "contents" : "",
+    "created" : 1557311882456.000,
+    "dirty" : false,
+    "encoding" : "UTF-8",
+    "folds" : "",
+    "hash" : "0",
+    "id" : "B6001EF7",
+    "lastKnownWriteTime" : 1557319656,
+    "last_content_update" : 1557319656,
+    "path" : "~/bmarketinggggg/bmarketing.R",
+    "project_path" : "bmarketing.R",
+    "properties" : {
+        "cursorPosition" : "51,0",
+        "scrollLine" : "28"
+    },
+    "read_only" : false,
+    "read_only_alternatives" : [
+    ],
+    "relative_order" : 1,
+    "source_on_save" : false,
+    "source_window" : "",
+    "type" : "r_source"
+}

.Rproj.user/E6F84AE4/sources/s-5EE927D0/B6001EF7-contents

@@ -0,0 +1,58 @@
+library(tidyverse)
+
+#################Loading data into the environment#################
+bmarketing <- read.csv2("bmarketing.csv")
+
+bmarketing$euribor3m = as.numeric(as.character(bmarketing$euribor3m))
+bmarketing$cons.conf.idx = as.numeric(as.character(bmarketing$cons.conf.idx))
+bmarketing$emp.var.rate = as.numeric(as.character(bmarketing$emp.var.rate))
+bmarketing$cons.price.idx = as.numeric(as.character(bmarketing$cons.price.idx))
+bmarketing$nr.employed = as.numeric(as.character(bmarketing$nr.employed))
+
+#Lets look at dataset and generate initial understanding about the column types
+str(bmarketing)
+summary(bmarketing)
+
+view (bmarketing)
+
+# A quick check:
+# If newdata has same number of observation that implies no NA value present
+# is.na(bmarketing)
+newdata <- na.omit(bmarketing)
+nrow(newdata)==nrow(bmarketing)
+
+#A deep check for a particular column let say age
+if(length(which(is.na(bmarketing$y)==TRUE)>0)){
+  print("Missing Value found in the specified column")
+} else{
+  print("All okay: No Missing Value found in the specified column")
+}
+
+# Let's find the range of individual variables
+summary(bmarketing)
+
+## ------------------------------------------------------------------------
+bmarketing %>% 
+  ggplot() + geom_histogram(aes(age), bins = 30) + 
+  geom_vline(aes(xintercept= median(age)), color = "red")
+
+# TODO: do boxplots for each data
+# boxplot(duration~y,data=bmarketing_sub,col="red")
+
+#################Decision Tree#################
+library(rpart)
+library(rpart.plot)
+
+dt_model<- rpart(y ~ ., data = bmarketing)
+# dt_model<- rpart(y ~ poutcome + emp.var.rate + cons.price.idx + cons.conf.idx + euribor3m + nr.employed, data = bmarketing)
+rpart.plot(dt_model)
+summary(dt_model)
+
+#################Testing Decision Tree #################
+predictions <- predict(dt_model, bmarketing, type = "class")
+
+## Compute the accuracy
+mean(bmarketing$y == predictions)
+
+# Lets look at the confusion matrix
+table(predictions, bmarketing$y)

.gitignore

@@ -0,0 +1,4 @@
+.Rproj.user
+.Rhistory
+.RData
+.Ruserdata

DESCRIPTION

@@ -0,0 +1,14 @@
+Package: Team1
+Type: Package
+Title: Data manage package
+Version: 0.1.0
+Date: 2019-05-08
+Author: Team 1 project 
+Maintainer: Hanna Popova <[email protected]>
+Imports: 
+  tidyverse,    
+  rpart,
+  rpart.plot
+Description: The package Team1 used for data cleaning, data transforming, model, model plotting, model predicting and model performance.
+License: open source
+RoxygenNote: 6.1.1

NAMESPACE

@@ -0,0 +1,15 @@
+# Generated by roxygen2: do not edit by hand
+
+export(clear1)
+export(clear2)
+export(clear3)
+export(histplot)
+export(intonum)
+export(logaritmic)
+export(model)
+export(model_performance)
+export(model_pred)
+export(treeplot)
+import(rpart)
+import(rpart.plot)
+import(tidyverse)

R/CLEAR.R

@@ -0,0 +1,98 @@
+#'Function to find missing data
+#'
+#'
+#'
+#'
+#'@param data Basic object like \code{numeric,int,boolean,char}
+#'@param target Basic object like \code{numeric,int,boolean,char}
+#'
+#'@keywords na.omit , nrow, length
+#'
+#'
+#'@export clear1
+#'@export clear2
+#'@export clear3
+#'
+#'@examples
+#'clear1(bmarketing$y)
+#'clear2(bmarketing$duration)
+#'clear3(bmarketing$poutcome)
+#
+#'
+
+
+
+
+# A quick check:
+# If newdata has same number of observation that implies no NA value present
+# is.na(bmarketing)
+
+clear1 <- function(data) {
+  newdata <- na.omit(data)
+  if (nrow(newdata)==nrow(data)) {
+    print("All okay: No Missing Value found")
+  } 
+  else{
+    print("Error: Missing Values found")
+  }
+}
+
+#'Function to find missing data inside a column
+#'
+#'
+#'
+#'
+#'@param data Basic object like \code{numeric,int,boolean,char}
+#'@param target Basic object like \code{numeric,int,boolean,char}
+#'
+#'@keywords na.omit , nrow, length
+#'
+#'
+#'@export clear2
+#'
+#'@examples
+#'clear2(bmarketing$duration)
+#
+#'
+
+#A deep check for a particular column let say age
+
+clear2 <- function(data,target) {
+  if(length(which(is.na(data$target)==TRUE)>0)){
+    print("Missing Value found in the specified column")
+  } 
+  else{
+    print("All okay: No Missing Value found in the specified column")
+  }
+}
+
+#'Remove any columns (and report as warning) which contain more than 50 percent NA’s
+#'
+#'
+#'
+#'
+#'@param data Basic object like \code{numeric,int,boolean,char}
+#'@param target Basic object like \code{numeric,int,boolean,char}
+#'
+#'@keywords na.omit, nrow, length
+#'
+#'
+#'@export clear3
+#'
+#'@examples
+#'clear3(bmarketing$duration)
+#
+#'
+
+#Remove any columns (and report as warning) which contain more than 50% NA’s
+clear3 <- function(data,target) {
+if(length(which(is.na(target)==TRUE)>length(target)/2)){
+  print("There are more than 50% of Missing Value in the specified column")
+  assign('target', NULL, envir = .GlobalEnv)
+
+} else{
+  print("All okay: There are no more than 50% of Missing Value in the specified column")
+}
+}
+
+

R/Plot.R

@@ -0,0 +1,38 @@
+
+#'Function to histogram
+#'
+#'
+#'
+#'
+#'@param data Basic object like \code{numeric, char, factor,boolean, NULL}
+#'@param column Basic object like \code{numeric}
+#'
+#'@keywords plot
+#'
+#'@import rpart.plot 
+#'@import rpart 
+#'@import tidyverse
+#'
+#'@export histplot
+#'@export treeplot
+#'
+#'@examples
+#'histplot(data, column)
+#'
+#'
+
+
+histplot<-function(data, column){
+
+  data %>% 
+    ggplot() + geom_histogram(aes(column), bins = 30) + 
+    geom_vline(aes(xintercept= median(column)), color = "red")
+}
+
+#trreplot
+
+treeplot <- function(model){
+
+  rpart.plot(model)
+
+}

R/Transform.R

@@ -0,0 +1,60 @@
+#'Transform numeric variables using the log 
+#'
+#'
+#'
+#'
+#'@param x Basic object like \code{numeric}
+#'@param data Basic object like \code{numeric, boolean, character, NULL, factor}
+#'@param column Basic object like \code{factor}
+#'
+#'@keywords log
+#'
+#'@import rpart.plot 
+#'@import rpart 
+#'@import tidyverse
+#'
+#'@export logaritmic
+#'
+#'@examples
+#'bmarketing$age<-logaritmic(bmarketing$age)
+#'bmarketing$duration<-logaritmic(bmarketing$duration)
+#'
+
+
+#################Transform numeric variables using the log #################
+
+logaritmic <- function(x) {
+
+x<-log(x)
+
+}
+
+#'Transform factors into numeric variables (and vice versa) as necessary
+#'
+#'
+#'
+#'
+#'@param data Basic object like \code{numeric, boolean, character, NULL, factor}
+#'@param column Basic object like \code{factor}
+#'
+#'@keywords number2factor
+#'
+#'@import rpart.plot 
+#'@import rpart 
+#'@import tidyverse
+#'
+#'@export intonum
+#'
+#'@examples
+#'bmarketing$poutcome<-intonum(bmarketing$poutcome)
+#'bmarketing$euribor3m<-intonum(bmarketing, bmarketing$euribor3m)
+#'
+
+#################Transform factors into numeric variables (and vice versa) as necessary#################
+
+intonum<-function(data, column) {
+
+data$column <- as.numeric(as.character(column))
+
+
+}

R/model.R

@@ -0,0 +1,26 @@
+#'Function to model
+#'
+#'
+#'
+#'
+#'@param dataset Basic object like \code{numeric, char, factor,boolean, NULL}
+#'
+#'@param target Basic object like \code{boolean}
+#'
+#'@keywords model, decision tree,
+#'
+#'@import rpart.plot 
+#'@import rpart 
+#'@import tidyverse
+#'
+#'@export model
+#'
+#'@examples
+#'model(bmarketing, bmarketing$y)
+#'
+#'
+
+# model function with 2 arguments: target and dataset 
+model <- function(dataset, target) {
+  rpart(target ~ ., data = dataset) 
+}

R/model_performance.R

@@ -0,0 +1,27 @@
+#'Function to model performance
+#'
+#'
+#'
+#'
+#'@param target Basic object like \code{boolean}
+#'
+#'@param predictions Basic object like \code{boolean}
+#'
+#'@keywords model, decision tree, accuracy
+#'
+#'@import rpart 
+#'@import tidyverse
+#'
+#'@export model_performance
+#'
+#'@examples
+#'model_performance(target, predictions)
+#'
+#'
+
+model_performance <- function(target, predictions) {
+  ## Compute the accuracy
+  mean(target == predictions)
+  # Lets look at the confusion matrix
+  table(predictions, target)
+}