r-在Shiny中，在给定复选框输入的情况下，如何对绘图的变量(x和或y)进行日志转换

Question

在Shiny中，如何在给定复选框输入的情况下对变量(x和或y(进行日志转换？如果输入变量是数字的，我希望在给定复选框输入的情况下对变量进行日志转换。

library(shiny)
library(ggplot2)
ui <- fluidPage(
selectInput(inputId = "xvariable",
label = "X Variable",
choices = colnames(mtcars)),
checkboxInput("LogX", "Log Transform", FALSE),
selectInput(inputId = "yvariable",
label = "Y Variable",
choices = colnames(mtcars)),
checkboxInput("LogY", "Log Transform", FALSE),

h3(""),
plotOutput("plot")
)
server <- function(input, output, session) {
output$plot <- renderPlot({
req(input$xvariable)
req(input$yvariable)
g <- ggplot(mtcars, aes(x = !!as.symbol(input$xvariable), y = !!as.symbol(input$yvariable)))
if (input$xvariable %in% c("mpg", "disp", "hp", "drat", "wt", "qsec")) {
# numeric
g <- g + geom_point()
} else {
# categorical
g <- g + geom_bar()
}
g
})
}
shinyApp(ui, server)

Answer 1

这应该可以工作。基于选定的变量，可以创建一个新的X和Y列。如果启用了日志转换，则转换该列。这也使ggplot调用更加容易。

library(shiny)
library(ggplot2)
ui <- fluidPage(
selectInput(inputId = "xvariable",
label = "X Variable",
choices = colnames(mtcars)),
checkboxInput("LogX", "Log Transform", FALSE),
selectInput(inputId = "yvariable",
label = "Y Variable",
choices = colnames(mtcars)),
checkboxInput("LogY", "Log Transform", FALSE),

h3(""),
plotOutput("plot")
)
server <- function(input, output, session) {
output$plot <- renderPlot({
req(input$xvariable)
req(input$yvariable)

## Copy mtcars into new dataframe
df <- mtcars

## Create Y Variable
# If LogY is TRUE
if(isTRUE(input$LogY)){
df$y <- log(df[[as.character(input$yvariable)]])
}
# If LogY is FALSE
if(!isTRUE(input$LogY)){
df$y <- df[[as.character(input$yvariable)]]
}
## Modify X Variable Based on Variable
if (input$xvariable %in% c("mpg", "disp", "hp", "drat", "wt", "qsec")) {
## Create X Variable
# If LogX is TRUE
if(isTRUE(input$LogX)){
df$x <- log(df[[as.character(input$xvariable)]])
}
# If LogX is FALSE
if(!isTRUE(input$LogX)){
df$x <- df[[as.character(input$xvariable)]]
}

## Graph
g <- ggplot(df, aes(x = x, y = y)) + geom_point()

} else {
df$x <- as.factor(df[[as.character(input$xvariable)]])

## Graph
g <- ggplot(df, aes(x = x)) + geom_bar()
}
g
})
}