在使用 gghighlight 时为几个 geom_hlines 创建图例

Posted 2023-02-16

【中文标题】在使用 gghighlight 时为几个 geom_hlines 创建图例

【英文标题】：Creating legends for several geom_hlines while using gghighlight

【发布时间】：2020-10-09 06:06:52

【问题描述】：

我想在 geom_hline 中为我的水平线创建一个图例。我使用的数据来自 3 个不同的数据框。我也在使用gghighlight，我相信它掩盖了任何传说。在这种情况下我如何强制传奇？

这 3 个数据框是： 1) 'dataset' 存储所有点的值 2) 'limits' 包含一些参数的最大值、最小值和目标 3) 'mean' 包含每个参数的平均值。

以下是数据集的最小可重复样本：

dataset <- data.frame(
  param = c('A','A','A','A','A', 'T','T','T','T','T', 'N','N','N','N','N', 'R','R','R','R','R'),
  category = c('Other','this','Other','Other','Other','this','Other','Other','Other','Other','Other','Other','this','Other','Other','Other','Other','Other','Other','this'),
  average = c(1.55,1.46,1.42,1.57,1.58, 1.57,1.46,1.42,1.57,1.59, 1.67,1.56,1.62,1.67,1.69, 1.47,1.36,1.32,1.47,1.49),
  datetime = c('2019-06-10 07:27:24','2019-06-10 08:20:24','2019-06-10 09:27:24','2019-06-10 07:45:24','2019-06-10 08:13:24',
               '2019-06-10 09:27:24','2019-06-10 10:20:24','2019-06-10 11:27:24','2019-06-10 09:45:24','2019-06-10 10:13:24',
               '2019-06-10 13:27:24','2019-06-10 14:20:24','2019-06-10 15:27:24','2019-06-10 13:45:24','2019-06-10 14:13:24',
               '2019-06-10 18:27:24','2019-06-10 19:20:24','2019-06-10 20:27:24','2019-06-10 18:45:24','2019-06-10 19:13:24')
)
dataset$datetime <- as.POSIXct(dataset$datetime, format = "%Y-%m-%d %H:%M:%S")

limits <- data.frame(
  param = c('A', 'T'),
  target = c(1.55, 1.55),
  min = c(1.39, 1.39),
  max = c(1.71, 1.71)
)

mean <- data.frame(
  param = c('A', 'T', 'N', 'R'),
  mean = c(1.549, 1.548, 1.65, 1.45)
)

这是我的代码：

library(ggplot2)
library(gghighlight)
ggplot(data=dataset, mapping=aes(x=datetime, y=average)) +
  geom_line(group=1, alpha=0.3, color='black') +
  geom_hline(data=limits, mapping=aes(yintercept = max), color='red', linetype='dashed') + #max line
  geom_hline(data=limits, mapping=aes(yintercept = min), color='red', linetype='dashed') + #min line
  geom_hline(data=limits, mapping=aes(yintercept = target), color='blue', linetype='dashed') + #target line
  geom_hline(data=mean, mapping=aes(yintercept = mean), color='green', linetype='dashed') + #mean line
  geom_point(size=2, color='red') +
  facet_wrap(param~., scales='free') +
  gghighlight(category!='Other',  label_key = average, n=1, use_group_by = FALSE,
              unhighlighted_params = list(color='black', size=1, alpha=0.7)) +
  labs(x='Time' , y='Value') +
  theme_bw()

图例应显示：“红色”：最小/最大，“蓝色”：目标，“绿色”：平均值。谢谢！

【参考方案1】：

您的代码稍作改动即可完成工作。

将颜色参数作为一种美学放在每个几何图形中，并将带有所需标签的字符串映射到它。

我没有使用gghighlight 来突出非其他点。出于某种原因，使用此功能禁用图例。

如果您需要更改颜色，只需按照每个几何图形的顺序添加和scale_color_manual()即可。

代码：

ggplot(data=dataset, mapping=aes(x=datetime, y=average)) +
    geom_line(group = 1, alpha = 0.3, color='black') +
    geom_hline(data=limits, mapping=aes(yintercept = max, color = "limits"), linetype='dashed') + #max line
    geom_hline(data=limits, mapping=aes(yintercept = min, color = "limits"), linetype='dashed') + #min line
    geom_hline(data=limits, mapping=aes(yintercept = target, color = "target"), linetype='dashed') + #target line
    geom_hline(data=mean, mapping=aes(yintercept = mean, color = "mean"), linetype='dashed') + #mean line
    geom_point(size=2) +
    geom_point(data = filter(dataset, category != "Other"),
               aes(x = datetime, y = average), color = "red", size = 3) +
    geom_label(data = filter(dataset, category != "Other"),
               aes(x = datetime, y = average, label = average), vjust = 1.5) +
    facet_wrap(param~., scales='free') +
    labs(x='Time' , y='Value') +
    theme_bw() +
    theme(legend.position = "bottom")