我想单独控制多面图中面板和绘图的背景颜色。在中使用提示 this answer ，我能够创建一些几乎可以工作的东西，但仍然有两个问题:

1. 颜色并不像我定义的那样出现 - 颜色因单元格而异，但颜色看起来“奇怪” - 因为它们是一些叠加层
2. 我无法单独控制面板和绘制背景颜色

到目前为止，这是我的代码:

ann_text <- data.frame(mpg = c(22, 15, 30), 
                       wt = c(4, 5, 2), 
                       cyl = factor(c(4, 6, 8), levels = c("4","6","8")),
                       lab = c("Text 1", "Text 2", "Text 3"),
                       hue = c("pink", "yellow", "lightblue"))
 
ggplot(mtcars, aes(mpg, wt)) +
  geom_rect(data = ann_text, aes(fill = hue),
            xmin = -Inf, xmax = Inf, ymin = -Inf, ymax = Inf, alpha = 0.3) +
  geom_text(data = ann_text, mapping = aes(x = mpg, y = wt, label = lab)) +
  theme(panel.background = element_blank(),
        panel.grid.major.y = element_line(color = "grey")) +
  facet_grid(. ~ cyl) +
  geom_point()

result of my code with 'Strange' colors

我本来想在 theme() 中使用类似于下面这行的内容，但是当然(？!)这行不通:

plot.background = element_rect(data = ann_text, mapping = aes(fill = hue))

有什么想法可以让我有机会指定面板和绘图的背景颜色？

最佳答案

使用scale_fill_identity。

ggplot(mtcars, aes(mpg, wt)) +
  geom_rect(data = ann_text, aes(fill = hue),
            xmin = -Inf, xmax = Inf, ymin = -Inf, ymax = Inf, alpha = 0.3) +
  geom_text(data = ann_text, mapping = aes(x = mpg, y = wt, label = lab)) +
  theme(panel.background = element_blank(),
        panel.grid.major.y = element_line(color = "grey")) +
  facet_grid(. ~ cyl) +
  geom_point() +
  scale_fill_identity()

或者，您可以将美学包装在 I

ggplot(mtcars, aes(mpg, wt)) +
  geom_rect(data = ann_text, aes(fill = I(hue)),
            xmin = -Inf, xmax = Inf, ymin = -Inf, ymax = Inf, alpha = 0.3) +
  geom_text(data = ann_text, mapping = aes(x = mpg, y = wt, label = lab)) +
  theme(panel.background = element_blank(),
        panel.grid.major.y = element_line(color = "grey")) +
  facet_grid(. ~ cyl) +
  geom_point()

https://stackoverflow.com/questions/64423420/