代码之家 › 专栏 › 技术社区 › tassones

ggplot2:更改轴文本的颜色以匹配组

ggplot2 r

1

tassones · 技术社区 · 2 年前

使用 tidyverse 和 mtcars 数据集,我正在绘制 model 和 mpg 每辆车的。我想要每辆车的轴文本颜色 模型 以反映 make (例如,丰田=红色,凯迪拉克=蓝色等)。我该怎么做?

示例数据集

library(tidyverse)

# Create make and model for each car
mtcars <- mtcars %>%
  rownames_to_column(var = 'car') %>%
  filter(!car %in% c('Mazda RX4 Wag', 'Hornet 4 Drive', 'Valiant', 'Ford Pantera L')) %>%
  separate_wider_delim(car, " ", names = c('make', 'model'), cols_remove = FALSE) %>%
  select(car, make, model, mpg)

head(mtcars)
#> # A tibble: 6 Ã 4
#>   car               make   model        mpg
#>   <chr>             <chr>  <chr>      <dbl>
#> 1 Mazda RX4         Mazda  RX4         21  
#> 2 Datsun 710        Datsun 710         22.8
#> 3 Hornet Sportabout Hornet Sportabout  18.7
#> 4 Duster 360        Duster 360         14.3
#> 5 Merc 240D         Merc   240D        24.4
#> 6 Merc 230          Merc   230         22.8

# How can I make the y-axis text color reflect the make of each model?
mtcars %>%
  ggplot(aes(x = mpg, y = model)) +
  geom_point() +
  labs(y = NULL) +
  scale_x_continuous(breaks = seq(0, 35, 5), limits = c(0,35)) +
  theme_bw()

^{创建于2023-12-06

reprex v2.0.2}

这里有一个类似的SO问题,它使用了一个只有两个类别的例子(这个例子有20个; link here )

1 回复 | 直到 2 年前

1

2

Allan Cameron 2 年前

官方不支持在主题元素中使用矢量化颜色,但它确实有效,因此您可以执行以下操作:

mtcars %>%
  ggplot(aes(x = mpg, y = model)) +
  geom_point() +
  labs(y = NULL) +
  scale_x_continuous(breaks = seq(0, 35, 5), limits = c(0,35)) +
  theme_bw() +
  theme(axis.text.y = element_text(
    color = scales::hue_pal()(length(levels(factor(mtcars$make))))[
      as.numeric(factor(mtcars$make))]
      ))

就我个人而言,我会避免使用颜色和图例作为轴标签。为什么不直接用facet标记呢?

mtcars %>%
  ggplot(aes(x = mpg, y = model)) +
  geom_point() +
  labs(y = NULL) +
  scale_x_continuous(breaks = seq(0, 35, 5), limits = c(0,35)) +
  theme_minimal() +
  theme(panel.spacing.y = unit(0, "mm"),
        strip.text.y.left = element_text(angle = 0),
        strip.background = element_rect(fill = "gray90", color = "white",
                                        linewidth = 1.5),
        strip.placement = "outside") +
  facet_grid(make~., scales = "free_y", space = "free_y", switch = "y")