I have developed dependant choicelist which works fine in my local (machine) whereas it's not working in Shinyapps.io. Please provide me suggestions.

karthees · July 20, 2019, 7:18pm

library(shiny)
library(shinymaterial)
library(flexdashboard)
library(dplyr)
library(tidyverse)
library(ggplot2)
library(plotly)
library(lime)
library(DT)
library(caret)
library(randomForest)

my_model <- base::readRDS("model/rf_default.rds")

customer scorecard inputs

main_vars <- c('Customer Name','Gender','Sector Code','Nationality','Account Type','Currency','Account Period in Month','Business Unit','Branch','Account Officer')
commercial_vars <- c('Product','Mobile Banking')
financial_vars <- c('Internet Banking','Total Charges')
customer_feature_vars <- c(main_vars,commercial_vars,financial_vars) %>% unique

Frequency_val <- c('Customer Demographics', 'Account Information', 'Bank Informantion', 'Financial Information')

base::load('data/dataset.RData')

churn_data_raw <- dataset.rf %>%
mutate(Numberoftransactions_bck =
dplyr::case_when (
No.OfDebitTrans < 3 ~ "<3k Trans",
No.OfDebitTrans < 6 ~ "3-5 Trans",
No.OfDebitTrans < 9 ~ "6-8 Trans",
No.OfDebitTrans < 15 ~ "9-14 Trans",
No.OfDebitTrans >= 15 ~ "Over 15 Trans",
TRUE ~ "NA"),
DebitTrans_Amt_bck = dplyr::case_when (
DebitTransAmt < 10000 ~ "<10k Trans.amt",
DebitTransAmt < 40000 ~ "10k-40k Trans.amt",
DebitTransAmt < 70000 ~ "40k-70k Trans.amt",
DebitTransAmt < 100000 ~ "70k-100k Trans.amt",
DebitTransAmt >= 10000 ~ "Over 100k Trans.amt",
TRUE ~ "NA"),
AvgBalance_bck = dplyr::case_when (
AvgBalance < 10000 ~ "<10k Avg.Balance",
AvgBalance < 180000 ~ "10k-180k Avg.Balance",
AvgBalance < 530000 ~ "180k-530k Avg.Balance",
AvgBalance >= 530000 ~ "Over 530k Avg.Balance",
TRUE ~ "NA"),
TotalCharges_bck = dplyr::case_when (
DebitTransAmt < 10000 ~ "<10k Charges",
DebitTransAmt < 40000 ~ "10k-40k Charges",
DebitTransAmt < 70000 ~ "40k-70k Charges",
DebitTransAmt < 100000 ~ "70k-100k Charges",
DebitTransAmt >= 100000 ~ "Over 100k Charges",
TRUE ~ "NA")
)

#tags$a(href='Mttutl',tags$img(src='side_nav1.png',height='60',width='90')),
ui <- material_page(
title = " ",
nav_bar_fixed = FALSE,
include_fonts = T,
includeCSS("www/css/material_css.css"),
nav_bar_color = "teal lighten-1",
material_side_nav(
image_source = "Vision_Excel.PNG",
material_side_nav_tabs(
side_nav_tabs = c(
"Customer Churn" = "example_side_nav_tab_1"
),
icons = c("person")
)
),
material_side_nav_tab_content(
side_nav_tab_id = "example_side_nav_tab_1",
material_tabs(
tabs = c(
"Customer Score" = "first_tab"
)
),
# Define tab content
material_tab_content(
tab_id = "first_tab",
material_row(
material_column(
width = 2,
material_card(
title = "",
depth = 6,
flexdashboard::gaugeOutput("churn_risk", width = "200px", height = "130px"),
tags$br(),
tags$br(),
sliderInput("range", "Churn Risk Range",min = 1, max = 100, value = c(70,100)),
material_dropdown(
input_id = "Branch",
label = "Branch",
choices = as.character(unique(churn_data_raw$Branch[order(churn_data_raw$predict.score[,2])])),
color = "#3c4858"
),

        material_dropdown(
          input_id = "account_Type",
          label = "Account Type",
          choices = as.character(unique(churn_data_raw$AccountType[order(churn_data_raw$predict.score[,2])])),
          color = "#3c4858"
        ),
        material_dropdown(
          input_id = "customername",
          label = "Customer Name",
          choices = as.character(unique(churn_data_raw$CustomerName[order(churn_data_raw$predict.score[,2])])),
          color = "#3c4858"
        ),
        material_dropdown(
          input_id = "accountno",
          label = "Account No",
          choices = as.character(unique(churn_data_raw$ContractID[order(churn_data_raw$predict.score[,2])])),
          color = "#3c4858"
        ),
        tags$br()
      )
    )
  )
)

)
)

server <- function(input, output, session) {

churn_analysis_data_branch <- reactive({

churn_data_filtered <- churn_data_raw %>%  dplyr::filter(churn_data_raw$predict.score[,2] >= input$range[1]/100 & churn_data_raw$predict.score[,2] <= input$range[2]/100 )

churn_data_filtered

})

observeEvent(
# c(
input$range,
# input$Branch,
# input$account_Type,
# input$customername,
# input$accountno
# ),
{
# if(input$range[1]< 0 | input$range[2]>100){

  if(is.null(input$range)){
    return(NULL)
  }
  
  churn_analysis_data_branch <- churn_analysis_data_branch()
  
  update_material_dropdown(session, "Branch",
                           value = as.character(unique(churn_analysis_data_branch$Branch[order(churn_analysis_data_branch$predict.score[,2])]))[1],
                           choices = as.character(unique(churn_analysis_data_branch$Branch[order(churn_analysis_data_branch$predict.score[,2])]))
  )
  
  churn_analysis_data_acctype <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch)
  
  
  update_material_dropdown(
    session,
    "account_Type",
    value = as.character(unique(churn_analysis_data_acctype$AccountType[order(churn_analysis_data_acctype$predict.score[,2])]))[1],
    choices = as.character(unique(churn_analysis_data_acctype$AccountType[order(churn_analysis_data_acctype$predict.score[,2])]))
  )
  
  churn_analysis_data_custname <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch & AccountType==input$account_Type)
  
  update_material_dropdown(
    session,
    "customername",
    value = as.character(unique(churn_analysis_data_custname$CustomerName[order(churn_analysis_data_custname$predict.score[,2])])[1]),
    choices = unique(churn_analysis_data_custname$CustomerName[order(churn_analysis_data_custname$predict.score[,2])])
  )
  
  churn_analysis_data_custname <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch & AccountType==input$account_Type & CustomerName==input$customername)
  
  update_material_dropdown(
    session,
    "accountno",
    value = as.character(unique(churn_analysis_data_custname$ContractID[order(churn_analysis_data_custname$predict.score[,2])])[1]),
    choices = unique(churn_analysis_data_custname$ContractID[order(churn_analysis_data_custname$predict.score[,2])])
  )
  
  
})

observeEvent(input$Branch, {

if(is.null(input$Branch)){
  return(NULL)
}

churn_analysis_data_acctype <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch)


update_material_dropdown(
  session,
  "account_Type",
  value = as.character(unique(churn_analysis_data_acctype$AccountType[order(churn_analysis_data_acctype$predict.score[,2])]))[1],
  choices = as.character(unique(churn_analysis_data_acctype$AccountType[order(churn_analysis_data_acctype$predict.score[,2])]))
)

churn_analysis_data_custname <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch & AccountType==input$account_Type)

update_material_dropdown(
  session,
  "customername",
  value = as.character(unique(churn_analysis_data_custname$CustomerName[order(churn_analysis_data_custname$predict.score[,2])])[1]),
  choices = unique(churn_analysis_data_custname$CustomerName[order(churn_analysis_data_custname$predict.score[,2])])
)

churn_analysis_data_custname <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch & AccountType==input$account_Type & CustomerName==input$customername)

update_material_dropdown(
  session,
  "accountno",
  value = as.character(unique(churn_analysis_data_custname$ContractID[order(churn_analysis_data_custname$predict.score[,2])])[1]),
  choices = unique(churn_analysis_data_custname$ContractID[order(churn_analysis_data_custname$predict.score[,2])])
)

})

observeEvent(input$account_Type, {
if(is.null(input$account_Type)){
return(NULL)
}

churn_analysis_data_custname <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch & AccountType==input$account_Type)

update_material_dropdown(
  session,
  "customername",
  value = as.character(unique(churn_analysis_data_custname$CustomerName[order(churn_analysis_data_custname$predict.score[,2])])[1]),
  choices = unique(churn_analysis_data_custname$CustomerName[order(churn_analysis_data_custname$predict.score[,2])])
)

churn_analysis_data_custname <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch & AccountType==input$account_Type & CustomerName==input$customername)

update_material_dropdown(
  session,
  "accountno",
  value = as.character(unique(churn_analysis_data_custname$ContractID[order(churn_analysis_data_custname$predict.score[,2])])[1]),
  choices = unique(churn_analysis_data_custname$ContractID[order(churn_analysis_data_custname$predict.score[,2])])
)

})

observeEvent(input$customername, {
if(is.null(input$customername == 0)){
return(NULL)
}

churn_analysis_data_custname <- dplyr::filter(churn_analysis_data_branch(),Branch == input$Branch & AccountType==input$account_Type & CustomerName==input$customername)

update_material_dropdown(
  session,
  "accountno",
  value = as.character(unique(churn_analysis_data_custname$ContractID[order(churn_analysis_data_custname$predict.score[,2])])[1]),
  choices = unique(churn_analysis_data_custname$ContractID[order(churn_analysis_data_custname$predict.score[,2])])
)

})

output$churn_risk <- renderGauge({

req(input$accountno)

selected_accountno <- churn_data_raw$ContractID[1]
selected_accountno <- input$accountno

# test_tbl_with_ids_predictions <- churn_data_raw %>%
#   mutate(churn_prob = predictions$Yes)

customer_tbl <- churn_data_raw %>% dplyr::filter(ContractID == selected_accountno)

p1  <- flexdashboard::gauge(
  round(customer_tbl$predict.score[,2]* 100, digits = 1),
  min = 0,
  max = 100,
  gaugeSectors(
    success = c(0,33),
    warning = c(33, 66),
    danger = c(67,100)
  ),
  symbol = "%",
  label = "Churn Risk"
)

})

}

shinyApp(ui = ui, server = server)

Gender	Married	AccountPeriodinMonth	MobileBanking	Nationality	Branch	InternetBanking	Currency	State	City	Product	SectorCode	ConsumerComplaints	BusinessUnit	AccountType	AccountOfficer	No.OfDebitTrans	DebitTransAmt	AvgBalance	TotalCharges	Churn	predict.class	predict.score.0	predict.score.1	CustomerID	ContractID	Country	LeBook	CustomerName
Female	Yes	1	No	KE	City Centre Branch	No	KES	Nairobi	Nairobi	CAA - CORPORATE	staff	No	Platinum Banking	CAA	JOB K KURGAT	0	0	0	29.85	0	0	0.506	0.494	1000001	1001	KE	1	AGUTA
Male	No	34	Yes	IN	City Centre Branch	Yes	GBP	TamilNadu	Chennai	SBA - NIC SAVER	Private Enter./other non-fin corp	No	SME	SBA	ERIC M KASAMBA	0	0	0	1889.5	0	0	0.994	0.006	1000001	1002	KE	1	AGUTA
Male	No	2	Yes	IN	City Centre Branch	Yes	KES	TamilNadu	Chennai	CAA - CORPORATE	staff	No	Platinum Banking	CAA	ERIC M KASAMBA	2	49320	561120	108.15	1	1	0.382	0.618	1000001	1003	KE	1	AGUTA
Male	No	45	No	KE	City Centre Branch	Yes	GBP	Nairobi	Nairobi	SBA - NIC SAVER	Private Enter./other non-fin corp	Yes	SME	SBA	JENIFFER W. NJENGA	4	60780	168397.5	1840.75	0	0	1	0	1000001	1004	KE	1	AGUTA
Female	No	2	Yes	IN	Junction Branch	No	GBP	TamilNadu	Chennai	CAA - CORPORATE	staff	No	Platinum Banking	CAA	JOB K KURGAT	0	0	808480	151.65	1	1	0.09	0.91	1000001	1005	KE	1	AGUTA
Female	No	8	Yes	US	Junction Branch	No	GBP	Dallas	Texas	CAA - CORPORATE	staff	No	Platinum Banking	CAA	JOB K KURGAT	0	0	379310	820.5	1	1	0.054	0.946

system · September 12, 2019, 11:18pm

This topic was automatically closed 54 days after the last reply. New replies are no longer allowed.