Shiny R：使用来自单选按钮的输入作为 by() 中的参数

Question

我有一个简单的单选按钮，允许用户选择“区域”或“房间大小”。我希望将输入传递给 by() 的 INDICES 参数，以便我可以为数据帧的每个因子级别运行一个函数并将结果呈现在一个表中。

我尝试过的几件事是：

 df <- Heating
 output$allmodelsbychar <- renderTable({    
    var <- input$runallmodels

    betas <- by(df, df$var, function(df){as.data.frame(summary(mlogit(depvar ~ ic + oc | 0, data= df, shape = "wide", choice = "depvar", varying = c(3:12)))$CoefTable)}, simplify = T)
    betas <- do.call("rbind", betas)
    betas
  })   
}

还有：

 df <- Heating
 output$allmodelsbychar <- renderTable({    

    betas <- by(df, df[ ,input$runallmodels], function(df){as.data.frame(summary(mlogit(depvar ~ ic + oc | 0, data= df, shape = "wide", choice = "depvar", varying = c(3:12)))$CoefTable)}, simplify = T)
    betas <- do.call("rbind", betas)
    betas
  })   
}

还有：

df <- Heating
 output$allmodelsbychar <- renderTable({

dfindecies <- reactive({
  if (input$runallmodels == "regiontype") {
    df <- df[, 15]
  }

  if (input$runallmodels == "roomsize") {
    df <- df[, 16]
    return(df)
  }
})

betas <- by(df, dfindecies(), function(df){as.data.frame(summary(mlogit(depvar ~ ic + oc | 0, data= df, shape = "wide", choice = "depvar", varying = c(3:12)))$CoefTable)}, simplify = T)
betas <- do.call("rbind", betas)
betas

Answer 1

我使用if eles解决了这个问题

server <- function(input, output) {

  df <- Heating
output$allmodelsbychar <- renderTable({

dfindecies <- reactive({
  if (input$runallmodels == "regiontype") {
    df <- df[, 16]
  } else {
    df <- df[, 15]
    return(df)
  }
})

### Do stuff

  }) 



}