r - 如何在 r 中包含分段 geom_smooth 的标签？

Question

我想在 r 中标记我的 geom_smooth，但标签更接近实际的点，而不是线。

数据:

df <- structure(list(t = c(45, 45, 45, 45, 45, 46, 46, 46, 46, 46, 
47, 47, 47, 47, 47, 48, 48, 48, 48, 48, 49, 49, 49, 49, 49, 50, 
50, 50, 50, 50, 51, 51, 51, 51, 51, 52, 52, 52, 52, 52, 53, 53, 
53, 53, 53, 54, 54, 54, 54, 54, 55, 55, 55, 55, 55, 56, 56, 56, 
56, 56, 57, 57, 57, 57, 57, 58, 58, 58, 58, 58, 59, 59, 59, 59, 
59, 60, 60, 60, 60, 60, 61, 61, 61, 61, 61, 62, 62, 62, 62, 62, 
63, 63, 63, 63, 63, 64, 64, 64, 64, 64, 65, 65, 65, 65, 65), 
    x = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
    0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1), name = c("P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4", "P1", "P2", "P2+", "P3", "P4", "P1", "P2", 
    "P2+", "P3", "P4"), value = c(48L, 132L, 111L, 115L, 2L, 
    58L, 126L, 82L, 74L, 0L, 45L, 119L, 78L, 87L, 0L, 56L, 106L, 
    105L, 88L, 1L, 52L, 78L, 91L, 107L, 1L, 35L, 96L, 86L, 98L, 
    1L, 61L, 118L, 90L, 108L, 2L, 45L, 114L, 93L, 98L, 2L, 55L, 
    108L, 78L, 76L, 7L, 44L, 97L, 94L, 96L, 0L, 40L, 111L, 93L, 
    88L, 1L, 43L, 78L, 66L, 113L, 2L, 20L, 57L, 84L, 41L, 0L, 
    17L, 51L, 81L, 34L, 0L, 40L, 55L, 64L, 32L, 0L, 25L, 67L, 
    71L, 37L, 0L, 16L, 67L, 60L, 57L, 0L, 23L, 46L, 62L, 47L, 
    1L, 34L, 75L, 68L, 39L, 0L, 34L, 60L, 85L, 24L, 0L, 20L, 
    58L, 63L, 37L, 1L)), row.names = c(NA, -105L), class = c("tbl_df", 
"tbl", "data.frame"))

我的代码:

df %>% 
  group_by(name) %>% 
  # New column to label the first and last values
  mutate(label =
           case_when(
             t == min(t) | t == max(t) ~ name,
             TRUE ~ NA_character_)) %>% 
  ggplot(aes(x = t, y = value, color = name, group = interaction(x, name), shape = name)) +
  geom_smooth(se = F) +
  geom_vline(xintercept = 57, linetype = "dashed", lwd = 0.3) + 
  scale_color_discrete(guide = 'none') +
  geom_label_repel(aes(label = label))

我的图表:

此外，是否可以去掉指向平滑黄土曲线的直线？

Answer 1

实现所需结果的一个选项是在 ggrepel::geom_label_repel 中使用 stat="smooth"。这样做会将标签放在拟合点上而不是数据点上。但是，要完成这项工作，您必须使用例如 geom_label_repel 动态添加标签after_stat(ifelse(x %in% range(x), color, NA_character_)):

library(ggplot2)
library(ggrepel)

ggplot(df, aes(x = t, y = value, color = name, group = interaction(x, name), shape = name)) +
  geom_smooth(se = F) +
  geom_vline(xintercept = 57, linetype = "dashed", lwd = 0.3) +
  scale_color_discrete(guide = "none") +
  ggrepel::geom_label_repel(aes(label = after_stat(ifelse(x %in% range(x), color, NA_character_))), stat = "smooth")

编辑 可能有不同的路线将标签放在数据范围(下)上端的左侧(右侧)。我的方法可能并不优雅。但是在我最近不得不准备很多图表的端点标签之后，我决定通过两个文本层来做到这一点在我看来是实现这一目标的最简单方法，因为它允许我们将标签向左移动或对。为了完成这项工作，我增加了 x 比例的扩展以为标签腾出空间。此外，我删除了编辑中的段并设置了 .direction = "y" 以便标签仅在 y 方向被排斥。

library(ggplot2)
library(ggrepel)

ggplot(df, aes(x = t, y = value, color = name, group = interaction(x, name), shape = name)) +
  geom_smooth(se = F) +
  geom_vline(xintercept = 57, linetype = "dashed", lwd = 0.3) +
  scale_color_discrete(guide = "none") +
  scale_x_continuous(expand = c(.1, .1)) +
  ggrepel::geom_label_repel(aes(label = after_stat(ifelse(x %in% range(x)[1], color, NA_character_))), 
                            stat = "smooth",
                            nudge_x = -1,
                            min.segment.length = Inf,
                            direction = "y") +
  ggrepel::geom_label_repel(aes(label = after_stat(ifelse(x %in% range(x)[2], color, NA_character_))), 
                            stat = "smooth",
                            nudge_x = 1,
                            min.segment.length = Inf,
                            direction = "y")