在函数中包含的ggplot2中使用重新排序

Question

背景

我正在构建一个函数来包装一些ggplot2，如下所示：

df <- data.frame("Variable" = c("CHK_ACCOUNT3", "AMOUNT", "DURATION"), "Overall_Imp" = c(71.44, 54.24, 34.84))

    clevelandPlot <- function (dataFrame, xValue, yValue) {

  ggplot(data = dataFrame, aes_string(x=xValue, y=yValue)) +
    geom_segment(aes_string(yend = yValue), xend = 0, colour = "grey50") +
    geom_point(size=3) +
    scale_x_continuous(expand = c(0.01, 0)) +
    theme_bw() +
    theme(panel.grid.major.y = element_blank()) +
    xlab("Importance") +
    ylab("")

}

此功能有效：

clevelandPlot(df, "Overall_Imp", "Variable")

但是当我尝试使用以下代码重新排序值时，它不起作用：

clevelandPlot <- function (dataFrame, xValue, yValue) {

  ggplot(data = dataFrame, aes_string(x=xValue, y=reorder(yValue, xValue))) +
    geom_segment(aes_string(yend = yValue), xend = 0, colour = "grey50") +
    geom_point(size=3) +
    scale_x_continuous(expand = c(0.01, 0)) +
    theme_bw() +
    theme(panel.grid.major.y = element_blank()) +
    xlab("Importance") +
    ylab("")

}

问题

如何按递减顺序对y轴中的变量进行重新排序（图表顶部的最大值）？

Answer 1

clevelandPlot <- function (dataFrame, xValue, yValue) {

  ix <- sort.int(dataFrame[,xValue], decreasing = FALSE, index.return = TRUE)$ix
  dataFrame[,yValue] <- factor(dataFrame[,yValue], dataFrame[ix,yValue])

  ggplot(data = dataFrame, aes_string(x=xValue, y=yValue)) +
    geom_segment(aes_string(yend = yValue), xend = 0, colour = "grey50") +
    geom_point(size=3) +
    scale_x_continuous(expand = c(0.01, 0)) +
    theme_bw() +
    theme(panel.grid.major.y = element_blank()) +
    xlab("Importance") +
    ylab("")
}

clevelandPlot(df, "Overall_Imp", "Variable")