i made a random forest predictive model and want my users to input the independent variables and predict the dependent variable. i'm using shiny to implement this.
i am getting the following error as the reactive values are stored as character inputs where as my predictive model is built with factor variables.
Warning: Error in predict.randomForest: Type of predictors in new data do not match that of the training data.
can any one help me how to convert this to a data frame with factor variables similar to my training data set. i used the following code.
i am getting the following error as the reactive values are stored as character inputs where as my predictive model is built with factor variables.
Warning: Error in predict.randomForest: Type of predictors in new data do not match that of the training data.
can any one help me how to convert this to a data frame with factor variables similar to my training data set. i used the following code.
Code:
#ui.R
library(shiny)
library(shinydashboard)
library(randomForest)
ui = dashboardPage(
dashboardHeader(title = "title"),
dashboardSidebar(
selectInput("transaction_type", "select transaction type:", choices =
tranlist),
selectInput("entry_type", "select entry type", choices = entrylist),
selectInput("sic", "select sic", choices = siclist),
selectInput("markettype", "select market type", choices = marketlist),
numericInput("qty", "enter quantity","",min = 0),
numericInput("volume", "enter Total_amnt","",min = 0),
actionButton("goButton","Enter")
),
dashboardBody(
tableOutput("result"),
textOutput("predict")
)
)
#server.R
function(input, output,session){
pos <- eventReactive(input$goButton,{
if(input$goButton >0){
test[1:7] = c(input$transaction_type, input$entry_type, input$qty, input$volume , input$sic, input$markettype, input$volume/input$qty)
} else{pos <- test}
test
})
output$result <- renderTable(pos())
output$predict = renderPrint(predict(settforest,newdata = pos(), type = "class"))
}