r - ggplot2 中 `geom_label` 的 y 值基于 `..count..` 变量

Question

我想创建一个直方图，其中有一条垂直线表示平均值，并在该线上附加一个标签，给出平均值的精确值。

我可以轻松创建带有垂直线的基本直方图。

# needed library
library(ggplot2)

# mean to be used later
x_mean <- mean(x = iris$Sepal.Length, na.rm = TRUE)

# creating basic plot with line for mean
(
  plot <- ggplot(data = iris,
                 mapping = aes(x = Sepal.Length)) +
    stat_bin(
      col = "black",
      alpha = 0.7,
      na.rm = TRUE,
      mapping = aes(y = ..count..,
                    fill = ..count..)
    )  +
    geom_vline(
      xintercept = x_mean,
      linetype = "dashed",
      color = "red",
      na.rm = TRUE
    ) +
    scale_fill_gradient(name = "count",
                        low = "white",
                        high = "white") +
    guides(fill = FALSE)
)
#> `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

现在我可以使用以下代码向该行添加标签:

# adding label to the line
plot +
  geom_label(mapping = aes(
    label = list(bquote("mean" == ~ .(
      format(round(x_mean, 2), nsmall = 2)
    ))),
    x = x_mean,
    y = 5  # how to automate this value choice?
  ),
  parse = TRUE)
#> `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

现在的问题是我正在对 y 进行硬编码-值 geom_label (y = 5)。这并不理想，因为如果我更改数据或变量或 binwidth，y = 5将不再是 y 轴的(大约)中间。我尝试设置 y = max(..count..)/2 ，但这会导致以下错误:

Error in FUN(X[[i]], ...) : object 'count' not found

总结一下:如何选择y值为 geom_label在这种情况下是否可以自动化，以便无论计数范围如何，标签始终以 Y 轴中间为中心？

Answer 1

您可以通过将代码中的硬编码 y = 5 替换为 y = Mean(layer_scales) 来从 plot 获取当前 y 轴范围(图)$y$range$range)。

这样，如果参数发生变化，则会考虑比例的变化。

# layer_scales(plot) gives the scale information for plot
> layer_scales(plot)$y
`stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
<ScaleContinuousPosition>
 Range:     0 --   12
 Limits:    0 --   12

# this is the actual vector for y-axis scale range
> layer_scales(plot)$y$range$range
`stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
[1]  0 12

# this is the y-axis midpoint value
> mean(layer_scales(plot)$y$range$range)
`stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
[1] 6