I'm looking at implementing 3D interactive plots in my shiny App, and so far I've been using plotly. However, plotly has one major disadvantage, it's extremely slow when rendering.
I've done checks, and the whole creation of updated outplot$plot <- renderPlotly ({.....}) and plotlyOutput("plot") takes less than 0.5 seconds, despite the large data set involved. This is a know issue for years but still seems to be current.
Hence, I'm looking to use a package called car, also because it has many options, some which I particularly want that are not available in other packages. Info on the car package is here: http://www.sthda.com/english/wiki/amazing-interactive-3d-scatter-plots-r-software-and-data-visualization
The problem is that it renders in a separate popup window rather than inside the shiny app, and I want to have it inside it, or even better, add a button that allow the user to have it as a popup, but only when asked. However, i can't figure out how to jam the bugger into the actual shiny page.
Here is my minimal example with a single text element and the graph code that (in my case) keeps appearing in a separate window rather than in the app.
install.packages(c("rgl", "car", "shiny"))
library("rgl")
library("car")
library(shiny)
cars$time <- cars$dist/cars$speed
ui <- fluidPage(
hr("how do we get the plot inside this app window rather than in a popup?"),
plotOutput("plot", width = 800, height = 600)
)
server <- (function(input, output) {
output$plot <- renderPlot({
scatter3d(x=cars$speed, y=cars$dist, z=cars$time, surface=FALSE, ellipsoid = TRUE)
})
})
shinyApp(ui = ui, server = server)
There is also this package, scatterplot3d but that's not interactive
http://www.sthda.com/english/wiki/scatterplot3d-3d-graphics-r-software-and-data-visualization
And there are some RGL packages but they have the same issue (seperate window) and don't offer the options I am lookign for.
You need to use an rglwidget which takes the last rgl plot and puts in in an htmlwidget. It used to be in a separate package, but it has recently been integrated into `rgl.
Here is the code to do this:
library(rgl)
library(car)
library(shiny)
cars$time <- cars$dist/cars$speed
ui <- fluidPage(
hr("how do we get the plot inside this app window rather than in a popup?"),
rglwidgetOutput("plot", width = 800, height = 600)
)
server <- (function(input, output) {
output$plot <- renderRglwidget({
rgl.open(useNULL=T)
scatter3d(x=cars$speed, y=cars$dist, z=cars$time, surface=FALSE, ellipsoid = TRUE)
rglwidget()
})
})
shinyApp(ui = ui, server = server)
Yielding this:
Related
I'm attempting to design a Shiny App where the user changes several inputs, then presses a "Plot" button to generate new visualizations. Currently, I have the a data object being generated in an eventReactive tied to the "Plot" button, with that data then getting fed into the renderPlot functions.
The framework seems to work, except that the plots will change whenever the inputs are changed. This often leads to errors in the plots, as different inputs load in different data. Pressing the "Plot" button after changing the inputs will cause the correct plots to render, but I'm trying to find a way to ensure that the plots don't change ever until that button is hit.
I believe the solution is a use of the "isolation" function, and I've tried that just about everywhere. However, none of these attempts have been successful. Below is a (simplified) setup of my code.
ui <- fluidPage(
sidebarLayout(
sidebarPanel(
# several selectInput options
actionButton('plot', label = 'Plot')
)
mainPanel(
plotOutput('outputPlot', height = '3in'),
)
)
)
server <- function(input, output) {
plotData <- eventReactive(input$plot, {
# load various data and organize into a list
return(data.list)
})
outPutPlot <- renderPlot({
plot.data <- plotData()
# manipulate data based on the specific plot, then generate
return(plot)
)
}
You’re right that you’d need to isolate() all reactive dependencies other
than plotData(). Without having a complete runnable example, it’s not
possible to point out where this might have gone wrong. However, it may be
easier to wrap the renderPlot() call in bindEvent(), which will isolate
everything. You just pass the expressions you want to depend on in other
arguments.
So try something like this:
bindEvent(renderPlot({ ... }), plotData())
I need to create shiny app which will create a plot basing on dropdown menu choise. The whole computation part is pretty complicated and so is the plot – I created a function which is returning ggplot and I just wanted to show it in the app.
My idea looks as follows:
library(shiny)
source('Analysis/function_external.R')
list_names = c('a', 'b', 'c')
ui <- fluidPage(
selectInput("data", "Select data to plot", choices = list_names)
)
# Define server logic required to draw a histogram
server <- function(input, output) {
observe({function_external(input$data)})
}
# Run the application
shinyApp(ui = ui, server = server)
It is making function run every time I change the input, but it does not show anything. I would really appreciate if you can point me into good direction.
output$my_complicated_plot <- renderPlot({ function_external(input$data) })
Solved the issue.
I am using shinycssloaders to show loading animation. There are multiple inputs on the page which are loaded from the server. These inputs are also dependent on each other.
In the below example I have used a reactive object to create such dependency. First the table is displayed and only when the calculation of table is completed (rv$a <- 1) plot can be completed.
library(shiny)
library(shinycssloaders)
ui <- fluidPage(
withSpinner(tableOutput('data')),
withSpinner(plotOutput('plot'))
)
server <- function(input, output) {
rv <- reactiveValues(a = NULL)
output$data <- renderTable({
#Some long calculation here, using Sys.sleep for simplicity
Sys.sleep(2)
rv$a <- 1
head(mtcars)
})
output$plot <- renderPlot({
req(rv$a)
#Some long calculation here, using Sys.sleep for simplicity
Sys.sleep(2)
plot(rnorm(100), rnorm(100))
})
}
shinyApp(ui, server)
This works fine but it shows 2 loaders, one for table and other one for plot. I want to combine these 2 loaders and show only 1 loading animation which covers the entire page combining table and plot. Also loading should end only after all the calculation is done i.e after plot calculation.
I have tried putting table and plot in a div and use spinner on div but it did not work and gave a blank page.
ui <- fluidPage(
withSpinner(div(
tableOutput('data'),
plotOutput('plot')
))
)
Does anybody have a solution to this? Is this possible using some different package?
You can use the tagList() function, which creates a list of the tags allowing the shinycssloaders package to wrap all the inputs within it in one go.
So the ui will look like the following:
ui <- fluidPage(
withSpinner(tagList(tableOutput('data'),
plotOutput('plot')))
)
Just some extra information for the animation. You can all change the style of the animation such as the the type, color, size , etc by adding these as arguments to the withSpinner(). Check the withSpinner() RDocumentation.
I'm building a shiny app that is supposed to show a bupaR process_map, which is kind of working.
Sadly the process_map()-function returns a dgr_graph-object, which can't be rendered with the DiagrammeR::renderGrViz() or DiagrammeR::renderDiagrammeR() functions. I found a way to convert it to a grViz / htmlwidget-object with render_graph(), but this way the graph is not nicely scaled. This is especially bad since I'm dealing with very long linear graphs.
Here is a MWE:
# server.R
library(DiagrammeR)
library(bupaR)
shinyServer(
function(input, output) {
plot <- patients %>%
process_map(rankdir = "TB",
render = FALSE)
output$diagram <- renderGrViz({
render_graph(plot)
})
}
)
# ui.R
library(DiagrammeR)
shinyUI(fluidPage(
titlePanel("DiagrammeR + shiny"),
grVizOutput(outputId = "diagram")
))
Is there a better way of displaying a dgr_graph-object in shiny, that fills out the whole width of the app and not just a post-stamp size in the middle? Ideal would be something adaptive, without the need for a fixed width in pixels.
As it turns out the problem is not the render_graph-function on the Server, but rather the default setting of the renderGrViz-function in the UI. When you set it to grVizOutput(outputId = "diagram", height = "100%") it works just fine.
I am trying to plot a simple histogram inside a shiny app. Outside the app (in R script), the graph gets plotted without any problems, (see here) but the same code produces an odd looking graph from inside the app (see the wrong graph here)
Could you help me figure out what's wrong? Link to dataset: https://drive.google.com/open?id=1ITK0lTWm_mkb9KonHLq4nuKDv-PBUDeR
library(ggplot2)
library(ggthemes)
library(scales)
library(shiny)
# read data
cso_corruption <- read.csv("cso_corruption.csv", header = T, sep = ",")
ui <- fluidPage(
sidebarLayout(
sidebarPanel(
radioButtons(inputId = "x",label ="Measurement Type",choices =
c("freq_business", "freq_cso"), selected = NULL)
),
mainPanel(
plotOutput(outputId = "hist")
)
)
)
server <- function(input, output) {
output$hist <- renderPlot(ggplot(data = na.omit(cso_corruption), aes(x=input$x)) +
geom_histogram(fill="lightgreen", stat="count"))
}
shinyApp(ui = ui, server = server)
Possible reason is the data is not making it into the server.
I would put the csv into another file within the directory, so you can access it using
read.csv("./Data/projectdata.csv")
Therefore it does not get lost when you publish. Also make sure the data is checked when publishing. One last note to include the read.csv function in the server.
After many trial-and-errors, I have figured out a solution. The trick is, in the server, I used aes_string instead of aes. I haven't figured out why aes_string works here, since it is supposed to require the variables to be passed in quotes. But it works for some reason.