This shiny app calculates sample size with given alpha and beta or power with provided alpha and sample size.

(Some codes were hided due to space limitation.)

Original shiny app : https://jihanyu.shinyapps.io/DDP_sample_num_power/

    ...
    sidebarLayout(
        sidebarPanel(
            h3("Statistical parameters"),
            sliderInput("a", "alpha:", min=0.001, max=1, value=0.05),
            sliderInput("b", "beta:", min=0, max=1, value=0.2),
            h3("Distribution parameters"),
            numericInput("mu0", "Mean of H0: ", value=0),
            numericInput("std.dev0", "Stand dev of H0: ", value=1),
            numericInput("std.dev1", "Stand dev of H1: ", value=1),
            numericInput("effect.size", "Effect size: ", value=0.5), 
            numericInput("n1", "sample number(H1): - if determined", value=NULL), 
            numericInput("n1.n0", "ratio of number(H1/H0): ", value=1), 
            checkboxInput("one.two", "2 sided test", value=TRUE),
            submitButton("Submit")
        ),
        mainPanel(
            plotOutput("dispPlot"), verbatimTextOutput("print.cal.sample.n"),
            verbatimTextOutput("print.prov.sample.n"), verbatimTextOutput("print.power")        )
    )

shinyServer(function(input, output) {
    ...
    two.sided <- reactive({
        if(input$one.two == TRUE){ 2 } else{ 1 }
    })
    q12 <- reactive({
        if(input$n1.n0 >= 1){ 1 + 1/input$n1.n0 } else{ 1 + input$n1.n0 }
    })
    est.sample.size <- reactive({
        ( (za() + zb())^2 * input$std.dev0^2 * q12() ) / (input$effect.size^2)
    })

    n1.new <- reactive({
        if(is.na(input$n1)){ est.sample.size() } else{ input$n1 }
    })
    n0 <- reactive({ n1.new()/input$n1.n0 })
    
    power.b <- reactive({
        pnorm( qnorm(input$a/two.sided(), mean=input$mu0, sd=std.err0(), lower.tail=FALSE),
               mean=mu1(), sd=std.err1(), lower.tail=FALSE)
    })
    ...

shinyServer(function(input, output) {
  ...
  output$dispPlot <- renderPlot({
        x <- seq(from = input$mu0-5*std.err0(), to=mu1()+5*std.err1(), by=0.01)
        h0 <- dnorm(x, mean=input$mu0, sd=std.err0())
        h1 <- dnorm(x, mean=mu1(), sd=std.err1())
        h.norm <- data.frame(x, h0, h1)
        a.h0 <- qnorm(input$a/two.sided(), mean=input$mu0, sd=std.err0(), lower.tail=FALSE)
        b.h1 <- qnorm(input$b, mean=mu1(), sd=std.err1())
        p <- ggplot(h.norm, aes(x=x, y=h0)) +
            geom_line(aes(x=x, y=h0), color="red") + geom_line(aes(x=x, y=h1), color="blue") +
            geom_vline(xintercept=a.h0) + geom_vline(xintercept=b.h1, linetype="dashed") +
            geom_ribbon(data=subset(h.norm, a.h0 <= x & x <= max(x)), 
                        aes(ymin=0, ymax=h0, fill="H0"), alpha=0.5, legend=FALSE) +
            geom_ribbon(data=subset(h.norm, min(x) <= x & x <= a.h0),
                        aes(ymin=0, ymax=h1, fill="H1"), alpha=0.5, legend=FALSE)
        return(p)
     })
  ...