I want to run rstan on Shiny GUI.
But I cannot success even if the following very simple Shiny code with rstan.
Someone has any idea?
library(shiny)
# Define UI for application that draws a histogram
ui <- fluidPage(
# Application title
titlePanel("Old Faithful Geyser Data"),
# Sidebar with a slider input for number of bins
sidebarLayout(
sidebarPanel(
sliderInput("bins",
"Number of bins:",
min = 1,
max = 50,
value = 30)
),
# Show a plot of the generated distribution
mainPanel(
plotOutput("distPlot")
)
)
)
# Define server logic required to draw a histogram
server <- function(input, output) {
output$distPlot <- renderPlot({
# generate bins based on input$bins from ui.R
x <- faithful[, 2]
bins <- seq(min(x), max(x), length.out = input$bins + 1)
model <- rstan::stan_model(
model_code = "parameters {real y;} model {y ~ normal(0,1);} "
)
fit <- rstan::sampling(model)
# draw the histogram with the specified number of bins
# hist(x, breaks = bins, col = 'darkgray', border = 'white')
rstan::stan_hist(fit)
})
}
# Run the application
shinyApp(ui = ui, server = server)
When rstan file foo.stan is compiled then foo.rds is created, so, I should upload together ? or it is meaningless?
My credit card is declined,.. I am not sure why but maybe my credit card is domestic only. so I cannot upgrade my plan for more than 1Gb.
I'm unfamiliar with Stan, but running your code (without Shiny) gives me the following errors/problems:
You want your graph to be dependent on an input slider I presume, but nowhere in your Stan code do you use the variable x or bins, so the model is not using any user-data at this point.
Creating the stan_model with your provided (easy) function takes a long time, is this normal? You can add the argument verbose = T to stan_model to see its intermediate steps
Finally, when I try and run the sampling method, I get an error:
> fit <- rstan::sampling(model)
Error in cpp_object_initializer(.self, .refClassDef, ...) :
could not find function "cpp_object_initializer"
failed to create the sampler; sampling not done
Try and create a piece of normal R code that works first, then we'll worry about the Shiny
You can find help for creating an example here:
I forget to attach library, so and I also use a bins for histogram.
Someone tells me that the amazon web server is required for free web application.
I am not sure but, someone told me that Stan compiling needs more RAM than R studio providing ram.
library(shiny);library(Rcpp);library(rstan);
# Define UI for application that draws a histogram
ui <- fluidPage(
# Application title
titlePanel("Old Faithful Geyser Data"),
# Sidebar with a slider input for number of bins
sidebarLayout(
sidebarPanel(
sliderInput("bins",
"Number of bins:",
min = 1,
max = 50,
value = 30)
),
# Show a plot of the generated distribution
mainPanel(
plotOutput("distPlot")
)
)
)
# Define server logic required to draw a histogram
server <- function(input, output) {
fit <- shiny::reactive({
model <- rstan::stan_model(
model_code = "parameters {real y;} model {y ~ normal(0,1);} "
)
fit <- rstan::sampling(model)
return(fit)
})
output$distPlot <- renderPlot({
# generate bins based on input$bins from ui.R
x <- faithful[, 2]
bins <- seq(min(x), max(x), length.out = input$bins + 1)
# draw the histogram with the specified number of bins
# hist(x, breaks = bins, col = 'darkgray', border = 'white')
rstan::stan_hist(fit(),bins=input$bins)
})
}
# Run the application
shinyApp(ui = ui, server = server)
