用共享轴在图形中排列面板

Question

我正试着用三个面板来做一个图形。我想删除轴标题和文本，无论哪两个面板相邻(见下面的例子)，并将一个图例放置在空的方面。

​

面板A和B应该有一个相同的y轴，所有三个面板都有相同的x轴(A和B基本上是完整数据集和子集的图，而C表示百分比变化)。我在这里看到了很多关于堆栈溢出的有用的帖子，但是没有人确切地解释了我想知道的事情。

我尝试: 1)用facet_wrap()将面板绘制为面，使用element_blank()控制轴限值(如解释的here)，并使用this函数替换图例：

# function to replace legend
library(lemon)

shift_legend2 <- function(p) {
  # ...
  # to grob
  gp <- ggplotGrob(p)
  facet.panels <- grep("^panel", gp[["layout"]][["name"]])
  empty.facet.panels <- sapply(facet.panels, function(i) "zeroGrob" %in% class(gp[["grobs"]][[i]]))
  empty.facet.panels <- facet.panels[empty.facet.panels]

  # establish name of empty panels
  empty.facet.panels <- gp[["layout"]][empty.facet.panels, ]
  names <- empty.facet.panels$name
  # example of names:
  #[1] "panel-3-2" "panel-3-3"

  # now we just need a simple call to reposition the legend
  reposition_legend(p, 'center', panel=names)
}

# add dummy data for element_blank()
dat <- dat %>%
  mutate(y_min = ifelse(figure == "pct_change", 0.8, 0),
         y_max = ifelse(figure == "pct_change", 1.6, 150))

# plot
plot1 <- ggplot(dat, aes(x = year, y = values)) + 
  geom_line(aes(col = class)) +
  theme_cowplot(12) +
  facet_wrap(~figure, nrow = 2, scales = "free_y") +
  geom_blank(aes(y = y_min)) +
  geom_blank(aes(y = y_max))

# rearrange legend
shift_legend2(plot1)

( 2)将地块分开，然后再安排。

a <- ggplot(dat[dat$figure == "d_complete",], aes(x = year, y = values)) + 
  geom_line(aes(col = class)) +
  theme_cowplot(12) +
  theme(legend.position = "none",
                      axis.title.x = element_blank(),
                      axis.text.x = element_blank())

b <- ggplot(dat[dat$figure == "d_subset",], aes(x = year, y = values)) + 
  geom_line(aes(col = class)) +
  theme_cowplot(12) + 
  theme(legend.position = "none",
                     axis.title.y = element_blank(),
                     axis.text.y = element_blank()) +
  ylim(c(0, 120)

c <- ggplot(dat[dat$figure == "pct_change",], aes(x = year, y = values)) + 
  geom_line(aes(col = class)) +
  theme_cowplot(12) +
  theme(legend.position = "none")

plot2 <- plot_grid(a, b, c, ncol=2, align = "hv", axis = "bl")

以下是研究结果：

​

现在的问题是：

• 在图1中，面板A的x轴线消失了，我不能为面板A和C放置不同的y轴图例，而B面板的y轴文本仍然在图2中的
• 中，图之间有非常丑陋的空白，我不知道如何添加传奇

。

谢谢你的帮忙!

数据：

dput(dat)
structure(list(class = c("AUN", "AUN", "AUN", "AUN", "AUN", "AUN", 
"AUN", "AUN", "AUN", "CHE", "CHE", "CHE", "CHE", "CHE", "CHE", 
"CHE", "CHE", "CHE"), year = c(0L, 0L, 0L, 5L, 5L, 5L, 10L, 10L, 
10L, 0L, 0L, 0L, 5L, 5L, 5L, 10L, 10L, 10L), figure = c("d_subset", 
"d_complete", "pct_change", "d_subset", "d_complete", "pct_change", 
"d_subset", "d_complete", "pct_change", "d_subset", "d_complete", 
"pct_change", "d_subset", "d_complete", "pct_change", "d_subset", 
"d_complete", "pct_change"), values = c(73, 99, 1, 84, 118, 1.19191919191919, 
80, 96, 0.96969696969697, 12, 16, 1, 19, 27, 1.6875, 16, 20, 
1.25), y_min = c(0, 0, 0.8, 0, 0, 0.8, 0, 0, 0.8, 0, 0, 0.8, 
0, 0, 0.8, 0, 0, 0.8), y_max = c(150, 150, 1.6, 150, 150, 1.6, 
150, 150, 1.6, 150, 150, 1.6, 150, 150, 1.6, 150, 150, 1.6)), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -18L), groups = structure(list(
    class = c("AUN", "CHE"), .rows = list(1:9, 10:18)), row.names = c(NA, 
-2L), class = c("tbl_df", "tbl", "data.frame"), .drop = TRUE))

Answer 1

我认为你在第二个选择中非常接近。您只需要修改a、b和c图的plot.margin，以减少可怕的空白的大小，然后使用ggdraw将图例添加为grob。下面是生成所需情节的完整代码。

library(tidyverse)
library(cowplot)

# Your original plots with the corresponding plot.margin
# Set the bottom margin to 0 to decrease the vertical gap between a and c plots
a <- ggplot(dat[dat$figure == "d_complete",], aes(x = year, y = values)) + 
   geom_line(aes(col = class)) +
   theme_cowplot(12) +
   theme(legend.position = "none",
         axis.title.x = element_blank(),
         axis.text.x = element_blank(),
         # modify the plot's margins: top, right, bottom, left
         plot.margin = unit(c(0.2,0,0,0), "cm"))

# Use same margins as plot a
b <- ggplot(dat[dat$figure == "d_subset",], aes(x = year, y = values)) + 
   geom_line(aes(col = class)) +
   theme_cowplot(12) + 
   theme(legend.position = "none",
         axis.title.y = element_blank(),
         axis.text.y = element_blank(),
         # modify the plot's margins: top, right, bottom, left
         plot.margin = unit(c(0.2,0,0,0), "cm")) +
   ylim(c(0, 120))

# Set top margin to -0.7 to decrease the vertical gap (between a and c)        
c <- ggplot(dat[dat$figure == "pct_change",], aes(x = year, y = values)) + 
      geom_line(aes(col = class)) +
      theme_cowplot(12) +
      theme(legend.position = "none",
            # modify the plot's margins: top, right, bottom, left
            plot.margin = unit(c(-0.7,0,0.2,0), "cm"))

# Construct the c plot with legend
temp_plot <- dat %>%
   filter(figure == "pct_change") %>%
   ggplot(aes(x = year, y = values)) + 
   geom_line(aes(col = class))

# get the plot's legend and transform it to a grob
legend <- as_grob(get_legend(temp_plot))

# Put a, b and c plots together
plot2 <- plot_grid(a, b, c, ncol=2, align = "hv", axis = "bl")

# Add the legend
plot3 <- ggdraw(plot2) +
   # set x and y position to draw the legend, usually requires playing with values between -1 and 1
   draw_grob(legend, x = 0.25, y = 0.05, width = 1, height = 0.5)

​

​