28

无论如何让grid.arrange()充当split.screen()?我想安排一张桌子直接位于图例下方。

#create histogram
my_hist<-ggplot(diamonds, aes(clarity, fill=cut)) + geom_bar()

#create inset table
my_table<- tableGrob(head(diamonds)[,1:3],gpar.coretext =gpar(fontsize=8),gpar.coltext=gpar(fontsize=8), gpar.rowtext=gpar(fontsize=8))

grid.arrange(my_hist,my_table, ncol=2)

产生:

在此处输入图像描述

但我希望它看起来大致像这样:

在此处输入图像描述

我试过 split.screen () 但它似乎不适用于 ggplot 类型的图形。有什么建议么?谢谢。

4

2 回答 2

27

Dickoa 的回答非常简洁。我的让你对元素有更多的控制。

my_hist <- ggplot(diamonds, aes(clarity, fill=cut)) + geom_bar()

#create inset table
my_table <- tableGrob(head(diamonds)[,1:3], gpar.coretext = gpar(fontsize=8), gpar.coltext=gpar(fontsize=8), gpar.rowtext=gpar(fontsize=8))

#Extract Legend
g_legend <- function(a.gplot){
  tmp <- ggplot_gtable(ggplot_build(a.gplot))
  leg <- which(sapply(tmp$grobs, function(x) x$name) == "guide-box")
  legend <- tmp$grobs[[leg]]
  return(legend)}

legend <- g_legend(my_hist)

#Create the viewports, push them, draw and go up
grid.newpage()
vp1 <- viewport(width = 0.75, height = 1, x = 0.375, y = .5)
vpleg <- viewport(width = 0.25, height = 0.5, x = 0.85, y = 0.75)
subvp <- viewport(width = 0.3, height = 0.3, x = 0.85, y = 0.25)
print(my_hist + opts(legend.position = "none"), vp = vp1)
upViewport(0)
pushViewport(vpleg)
grid.draw(legend)
#Make the new viewport active and draw
upViewport(0)
pushViewport(subvp)
grid.draw(my_table)

在此处输入图像描述

于 2012-08-09T14:49:41.603 回答
15

首先你应该看一下这个Wiki,有很多例子(看一下arrangeGrob 一个)。所以使用这些例子,我设法得到了这个解决方案

require(gridExtra)
require(ggplot2)

## original graph
my_hist <- ggplot(diamonds, aes(clarity, fill=cut)) + geom_bar()

## get the legend 
tmp <- ggplot_gtable(ggplot_build(my_hist))
leg <- which(sapply(tmp$grobs, function(x) x$name) ==  "guide-box")
legend <- tmp$grobs[[leg]]

## create inset table
my_table <- tableGrob(head(diamonds)[,1:3],gpar.coretext =gpar(fontsize=8),gpar.coltext=gpar(fontsize=8), gpar.rowtext=gpar(fontsize=8))


### final result 
grid.arrange(my_hist + opts(legend.position = "none"), arrangeGrob(legend, my_table), ncol = 2)

在此处输入图像描述

于 2012-08-09T14:06:36.180 回答