Showing Model Predictions with Shiny in R - r

Question

I'd be happy to get some help with R shiny, I'm new to it.
I built a logistic regression model that runs on a training set called "Rdata1.csv".
Then after I train the model I want let the user upload a test set file
and run the predictions on the test set.
Here is my ui.R:
library(shiny)
shinyUI(fluidPage(
titlePanel("Predictions"),
sidebarLayout(
sidebarPanel(
fileInput("file", label = h3("Upload CSV")),
hr(),
fluidRow(column(4, verbatimTextOutput("value")))
),
mainPanel(
textOutput("text1")
)
)
))
Here is my server.R:
library(shiny)
library(aod)
library(ROCR)
mydata <- read.csv("C:/Rdata1.csv")
mydata$Continent <- factor(mydata$Continent)
mydata$IP <- factor(mydata$IP)
mylogit <- glm(Good ~ Continent + IP, data = mydata, family = "binomial")
shinyServer(function(input, output) {
output$text1 <- renderPrint({
mydatanew <- read.csv(input$file[4])
mydatanew$Continent <- factor(mydatanew$Continent)
mydatanew$IP <- factor(mydatanew$IP)
mydatanew$predicted<-predict(mylogit, newdata=mydatanew, type="response")
paste("predictions", mydatanew)
})
}
)
In the app I get no output - instead it I get the error:
"file must be a character string or connection" and i see nothing.
Anyone know what's wrong with the code?
Thanks!

Answer 1

Welcome to SO. This question is hard to make reproducible, but it looks like the datapath from fileInput is not being passed correctly. See ?fileInput:
Whenever a file upload completes, the corresponding input variable is set to a dataframe.
This dataframe contains one row for each selected file, and the following columns: ...
... the fourth of which is datapath.
In other words
file <- data.frame(name="foo",size=999,type="text/plain",datapath = "~/temp/whatever.csv", stringsAsFactors=FALSE)
str(file[4]) # this gives you a little data frame, probably not what you want
str(file[1,4]) # this gives you the character value itself
Does that work?