从ggrepel获取标签位置的坐标 [英] Getting coordinates for the label locations from ggrepel

问题描述

以下是使用ggplot2和ggrepel进行标签位置优化"的示例:

  x = c(0.8846，1.1554，0.9317，0.9703，0.9053，0.9454，1.0146，0.9012，0.9055、1.3307)y = c(0.9828，1.0329，0.931，1.3794，0.9273，0.9605，1.0259，0.9542，0.9717、0.9357)ShortSci = c("MotAlb"，"PruMod"，"EriRub"，"LusMeg"，"PhoOch"，"PhoPho"，"SaxRub"，"TurMer"，"TurPil"，"TurPhi")df <-data.frame(x = x，y = y，z = ShortSci)库(ggplot2)图书馆(ggrepel)ggplot(数据= df，aes(x = x，y = y))+ theme_bw()+geom_text_repel(aes(label = z)，box.padding = unit(0.45，"lines"))+geom_point(颜色=绿色"，大小= 3) 

问题:是否有可能以某种方式将标签位置的坐标作为矢量?

非常感谢！

解决方案

在@Henrik建议的链接上，有许多有趣的想法可用来获取 ggrepel 标签的位置.
这是个人返工:

  x = c(0.8846，1.1554，0.9317，0.9703，0.9053，0.9454，1.0146，0.9012，0.9055，1.3307)y = c(0.9828，1.0329，0.931，1.3794，0.9273，0.9605，1.0259，0.9542，0.9717，0.9357)ShortSci = c("MotAlb"，"PruMod"，"EriRub"，"LusMeg"，"PhoOch"，"PhoPho"，"SaxRub"，"TurMer"，"TurPil"，"TurPhi")df <-data.frame(x = x，y = y，z = ShortSci)库(ggplot2)图书馆(ggrepel)p1<-ggplot(数据= df，aes(x = x，y = y))+ theme_bw()+geom_text_repel(aes(label = z)，box.padding = unit(0.45，"lines"))+geom_point(颜色=绿色"，大小= 3)x11(宽度= 7，高度= 7)1#获取x和y的绘图范围xrg<-ggplot_build(p1)$ layout $ panel_ranges [[1]] $ x.rangeyrg<-ggplot_build(p1)$ layout $ panel_ranges [[1]] $ y.range图书馆(网格)grid.force()孩子<-childNames(grid.get("textrepeltree"，grep = TRUE))#功能:获取单个ggrepel标签的x和y位置get.xy.pos.labs<-function(n){grb<-grid.get(n)data.frame(x = xrg [1] + diff(xrg)* convertX(grb $ x，"native"，valueOnly = TRUE)，y = yrg [1] + diff(yrg)* convertY(grb $ y，"native"，valueOnly = TRUE))}#获取所有ggrepel标签的位置dts<-do.call(rbind，lapply(kids，get.xy.pos.labs))dts $ lab<-df $ z打印(DTS)#x y实验室#1 0.8871681 0.9945523 MotAlb#2 1.1441033 1.0214568 PruMod#3 0.9431081 0.9194218 EriRub#4 0.9977394 1.3906067 LusMeg#5 0.8900197 0.9160821照片#6 0.9597605 0.9471191 PhoPho#7 1.0033184 1.0144658 SaxRub#8 0.8898377 0.9424059 TurMer#9 0.9340703 0.9722663 TurPil#10 1.3063754 0.9358787 TurPhi#使用通过get.xy.pos.labs计算出的位置来绘制标签p2<-ggplot(数据= df，aes(x = x，y = y))+ theme_bw()+geom_text(aes(x = x，y = y，label = lab)，data = dts，col ="red")+geom_point(颜色=绿色"，大小= 3)x11(宽度= 7，高度= 7)2 

更新2018年12月
对于 3.1.0 (及更高版本)的 ggplot2 版本，您需要使用以下代码来获取x和y范围(请参阅@VilmantasGegzna的答案

Below is an example of "label location optimization" using ggplot2 and ggrepel:

x = c(0.8846, 1.1554, 0.9317, 0.9703, 0.9053, 0.9454, 1.0146, 0.9012, 
  0.9055, 1.3307)
y = c(0.9828, 1.0329, 0.931, 1.3794, 0.9273, 0.9605, 1.0259, 0.9542, 
  0.9717, 0.9357)
ShortSci = c("MotAlb", "PruMod", "EriRub", "LusMeg", "PhoOch", "PhoPho", 
         "SaxRub", "TurMer", "TurPil", "TurPhi")

df <- data.frame(x = x, y = y, z = ShortSci)
library(ggplot2)
library(ggrepel)
ggplot(data = df, aes(x = x, y = y)) + theme_bw() + 
  geom_text_repel(aes(label = z), 
                  box.padding = unit(0.45, "lines")) +
  geom_point(colour = "green", size = 3)

Question: is it at all possible to somehow grab the coordinates for the label locations as vectors?

Thank you very much!

解决方案

At the link suggested by @Henrik there many interesting ideas for getting ggrepel label locations.
Here is a personal reworking:

x = c(0.8846, 1.1554, 0.9317, 0.9703, 0.9053, 0.9454, 1.0146, 0.9012, 0.9055, 1.3307)
y = c(0.9828, 1.0329, 0.931, 1.3794, 0.9273, 0.9605, 1.0259, 0.9542, 0.9717, 0.9357)
ShortSci = c("MotAlb", "PruMod", "EriRub", "LusMeg", "PhoOch", "PhoPho", 
         "SaxRub", "TurMer", "TurPil", "TurPhi")
df <- data.frame(x = x, y = y, z = ShortSci)

library(ggplot2)
library(ggrepel)
p1 <- ggplot(data = df, aes(x = x, y = y)) + theme_bw() + 
  geom_text_repel(aes(label = z), 
                  box.padding = unit(0.45, "lines")) +
  geom_point(colour = "green", size = 3)
x11(width=7,height=7)
p1

# Get x and y plot ranges 
xrg <- ggplot_build(p1)$layout$panel_ranges[[1]]$x.range
yrg <- ggplot_build(p1)$layout$panel_ranges[[1]]$y.range

library(grid)
grid.force()
kids <- childNames(grid.get("textrepeltree", grep = TRUE))

# Function: get the x and y positions of a single ggrepel label
get.xy.pos.labs <- function(n) {
  grb <- grid.get(n)
  data.frame(
  x = xrg[1]+diff(xrg)*convertX(grb$x, "native", valueOnly = TRUE),
  y = yrg[1]+diff(yrg)*convertY(grb$y, "native", valueOnly = TRUE)
  )
}
# Get positions of all ggrepel labels
dts <- do.call(rbind, lapply(kids, get.xy.pos.labs))
dts$lab <- df$z
print(dts)

#            x         y    lab
# 1  0.8871681 0.9945523 MotAlb
# 2  1.1441033 1.0214568 PruMod
# 3  0.9431081 0.9194218 EriRub
# 4  0.9977394 1.3906067 LusMeg
# 5  0.8900197 0.9160821 PhoOch
# 6  0.9597605 0.9471191 PhoPho
# 7  1.0033184 1.0144658 SaxRub
# 8  0.8898377 0.9424059 TurMer
# 9  0.9340703 0.9722663 TurPil
# 10 1.3063754 0.9358787 TurPhi

# Plot labels using the positions calculated using get.xy.pos.labs      
p2 <- ggplot(data = df, aes(x = x, y = y)) + theme_bw() + 
  geom_text(aes(x=x,y=y, label=lab), data=dts, col="red")+
  geom_point(colour = "green", size = 3)
x11(width=7,height=7)
p2

UPDATE December 2018
With ggplot2 version 3.1.0 (and newer) you need to use the following code to get the x- and y-range (see the answer of @VilmantasGegzna here for details):

# Get x and y plot ranges 
ggp1 <- ggplot_build(p1)
xrg <- ggp1$layout$panel_params[[1]]$x.range
yrg <- ggp1$layout$panel_params[[1]]$y.range

Here are the plot with labels given by geom_text_repel and the plot with the labels positioned using the calculated positions.

这篇关于从ggrepel获取标签位置的坐标的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！