Show Prediction on Shiny Dashboard

[Tams101]

Hi awesome community, I am trying to build a shiny App that predicts theft (Electricity theft), but I am finding it difficult to create a dashboard according to my mind specification. I have created the dashboard but how do I:

1. Set the predicted data to the dashboard such that whenever data is uploaded and the prediction button is clicked, the output should be shown in the dashboard. The output here is # OF PREDICTED THEFT CASES, # OF NO THEFT CASES and TOTAL NUMBER OF CUSTOMERS (based on the uploaded data)
2. The dashboard should default to zero when no data is uploaded.

Here's my server code

library(caret)
#> Loading required package: lattice
#> Loading required package: ggplot2
library(shiny)
library(LiblineaR)
library(readr)
library(ggplot2)
library(tidyverse)
library(tidymodels)

model <- load("treemodel.rda")
source("C:/Users/PHED KEYACCTS/Documents/Theft Project/NEW UI/TheftApp/TheftApp/UI.R/Server/app.R")
server <- shinyServer(function(input, output) {
#####################********************#################################
options(shiny.maxRequestSize = 800*1024^2)     
    # The # of "views"
    output$Total <- renderValueBox({
        valueBox(
            value = format(totalAccounts, big.mark=","),
            subtitle = "Total # of Accounts",
            color = "aqua" 
        )
    })
    
    # The # of "reads"
    output$Bypass <- renderValueBox({
        valueBox(
            value = format(Bypass, big.mark=","),
            subtitle = "Total # of Bypass"
        )
    })
    
    # The # of "fans"
    output$NoBypass <- renderValueBox({
        valueBox(
            value = format(NoBypass, big.mark=","),
            subtitle = "Total # of No Bypass"
        )
    })

#############################********################************################
   predictions<-reactive({
        
        inFile <- input$file1
        
        if (is.null(inFile)){
            return(NULL)
        }else{
            withProgress(message = 'Predictions in progress. Please wait ...', {
                input_data =  readr::read_csv(input$file1$datapath, col_names = TRUE)
                
                colnames(input_data) = c("Status","Availability", "Average.Consumption", "Estimated.Connected.Load",
                                         "group_cons","MeterBypass","FeederBypass","TariffBypass")
                
                input_data$Status = as.factor(input_data$Status )
                
                levels(input_data$Status) <- c("Bypassed","NoBypass")
                prediction = predict(tree.mod,input_data)
                
                totalAccounts = sum(prediction)
                Bypass = sum(prediction == 1)
                NoBypass  = sum(prediction == 0)
                
                input_data_with_prediction = cbind(input_data,prediction )
                input_data_with_prediction
            })
        }
    })
#############################********################************################
    
    #Downlaod MenuItem
    output$sample_predictions = renderTable({   # the last 6 rows to show
        pred = predictions()
        head(pred)
    })
 
   # output$plot_predictions = renderPlot({   # the last 6 rows to show
    #pred = predictions()
    #cols <- c("Bypass" = "red","NoBypass" = "blue")
    #ggplot(pred, aes(x =status,y= value)+ geom_point(size = 4, shape = 19, alpha = 0.6) +
    #scale_colour_manual(values = cols,labels = c("Bypass", "NoBypass"),name="Test Result"))
    
    #})
    
    # Downloadable csv of predictions ----
    output$downloadData <- downloadHandler(
        filename = function() {
            paste("input_data_with_predictions", ".csv", sep = "")
        },
        content = function(file) {
            write.csv(predictions(), file, row.names = FALSE)
        })
    
})

#shinyApp(ui = ui, server = server)

...

[nirgrahamuk]

fileInput and actionButton will likely be needed
Chapter 3 Basic UI | Mastering Shiny (mastering-shiny.org)

[Tams101]

Thanks @nirgrahamuk

Yes, I have File input but no Action button.

Now, assuming I have all of those what do I need to do to get my pain point working??