Subscribe to DSC Newsletter

Ever since I’ve started working on R , I always wondered how I can present the results of my statistical models as web applications. After doing some research over the internet I’ve come across ShinyR – a new package

from RStudio which can be used to develop interactive web applications with R.

Before going into how to build web apps using R, let me give you some overview about ShinyR.

Features:

  • No JavaScript/HTML knowledge required. 
  • Can build interactive web applications which can load the data dynamically.
  • Pre-Built output widgets for displaying plots, tables interactively are available.
  • Works in any R environment.
  • Shiny Apps can be hosted locally/on self-hosted Shiny Server/Rstudio-hosted                              Shiny Server – in today’s post we will see hosting locally.

 Building a shiny App

Shiny application have two components: a user-interface definition and server script,  and any additional data, scripts, or other resources required to support the application.

UI script (ui.R):

In this script we define the UI layout of the webpage. We can make  use of inbuilt UI controls/widgets for displaying output and for creating UI layout.

Server Script (server.R): In this script, the actual business logic/ code for statistical models, plotting the output is defined. It accepts inputs  from UI layer and compute outputs send back the response to the UI layer. The basic task of a Shiny server script is to define the relationship between inputs and outputs.

Understanding UI.R script:

library(shiny) #load shiny package

ds = read.csv("inputdata.csv") #load data

x = names(ds) # Define UI for application that plots distributions

shinyUI(pageWithSidebar(

# Application title

headerPanel("Demo Shiny Application"),

#sidebar panel to the left side of the page for input controls, slider & dropdown

sidebarPanel(

sliderInput('Year', 'Year', min=2010, max=2012,

value=min(2010,2012), step=1, round=0),

selectInput('select', 'Select Criteria', c(x[3],x[9],x[10],x[11],x[12],x[26]))

),

#main panel for displaying outputs, Histograms, summary

mainPanel(

plotOutput('plot'),

verbatimTextOutput("summary")

 ) ))

 

Notice in particular, in the above code sidebarPanel and mainPanel functions  are called with two arguments - corresponding to the two inputs and  two outputs displayed.

In the above image we can see the layout which contains a slider with year range, where based on the year selection the dataset will be taken into consideration. Dropdownlist containing varaibles, placeholder to display the graphs, table to display the summaries of the models.

Understanding Server.R

In the above script, we can observe that all the logic for the exploratory analysis is written in server.R script. It contains a main function called ShinyServer() wherein we can write reactive functions,reactive plots, outputs. The outputs are send back to the UI layer. 

In the above code, I have written logic which contains data fetching, modification, setting the dataset based on the year selected in the slider (reactive function), displaying histogram based on the selection 

on slider and dropdownlist,writing summary to a table. 

Deploying the application:

Once we are done with the application, next step is to deploy the application. 

Using Shiny, we have two different options to move the application to Production.

  • Create a local- self-hosted Shiny Server on Linux platform, here.
  • Host on the R-Studio's Shiny server, explained here

Deploying locally:

 

As mentioned in the beginning of the post, in order to deploy locally, we make

use of GIST. The steps are quite simple:

  • Upload the Server.R & UI.R files to the Gist.Github.com website.
  • Uploading the files gives you an ID - in my case - I got "8942533"
  • After finishing upload, open R and run the below code:

               library('shiny')

               shiny::runGist('8942533')

  • The model will be displayed as web page in the browser with inputs & outputs.

Please go through more posts @ www.dataperspective.info

Views: 5552

Tags: R, shinyR

Comment

You need to be a member of AnalyticBridge to add comments!

Join AnalyticBridge

On Data Science Central

© 2019   AnalyticBridge.com is a subsidiary and dedicated channel of Data Science Central LLC   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service