r - 在 geom_line 上显示数据值

Question

我使用 ggplot 和 geom_line 创建了一个绘图，并希望在绘图上显示单个数据值(数字)。

这是我目前的代码:

ggplot(errors_prob_3) +

geom_point(aes(x = ID, y = train_mean, colour = 'red')) +
geom_line(aes(x = ID, y = train_mean, colour = 'red')) +

geom_point(aes(x = ID, y = train_sd, colour = 'blue')) +
geom_line(aes(x = ID, y = train_sd, colour = 'blue')) +

geom_point(aes(x = ID, y = test_mean, colour = 'green')) +
geom_line(aes(x = ID, y = test_mean, colour = 'green')) +

geom_point(aes(x = ID, y = test_sd, colour = 'purple') )+
geom_line(aes(x = ID, y = test_sd, colour = 'purple')) +

labs(x = "Number of predictors", y = "Mean/sd value", 
    title = "Plot of the training and test cv error") +
scale_color_discrete(name = "Legend", labels = c("Train mean", "Train sd", "Test mean", "Test sd")) 

我得到一个错误:

Aesthetics must be either length 1 or the same as the data (6): label.

errors_prob_3 是一个包含 6 个 obs 和 5 个变量的数据框。

我的数据如下:

structure(list(ID = 1:6, train_mean = c(9.62056569110574, 9.16711345988648, 7.6586555534977, 7.62016661214069, 7.56713995100505,7.30900730515059),
train_sd = c(0.986375934334889, 1.00056316593354, 0.710802251192404, 0.691415430116925, 0.687478712185106, 0.726304511632279), 
test_mean = c(10.3738135132565, 10.0369765520266, 8.14168965211194, 8.40469077569391, 8.23953878513202, 8.0334726272832), 
test_sd = c(3.78543823754458, 3.6662904556102, 2.89196938350374, 2.74691446446284, 2.76568576521599, 2.98708614111826)), 
row.names = c(NA, -6L), class = "data.frame")

我现在的剧情

任何帮助，将不胜感激。谢谢!

Answer 1

主要技巧是reshape your data from wide to long format .然后一切都变得更加简单甚至直观。

library(tidyverse)
library(ggplot2)

errors_prob_3 %>%
  gather(key = "variable", value = "value", -ID) %>%
  ggplot(aes(x = ID, y = value, colour = variable)) +
  geom_line() +
  geom_point() +
  geom_text(aes(label = round(value, 1)),
            vjust = "inward", hjust = "inward",
            show.legend = FALSE) +
  labs(x = "Number of predictors", y = "Mean/sd value", 
       title = "Plot of the training and test cv error") +
  scale_color_discrete(name = "Legend", labels = c("Train mean", "Train sd", "Test mean", "Test sd"))

---

编辑

要仅显示第一个或最后一个文本标签，请对数据进行子集化。在这种情况下，我将显示每行的最后一个标签。

errors_prob_3 %>%
  gather(key = "variable", value = "value", -ID) %>%
  ggplot(aes(x = ID, y = value, colour = variable)) +
  geom_line() +
  geom_point() +
  geom_text(
    data = . %>% filter(ID == max(ID)),
    aes(label = round(value, 1)),
    vjust = "inward", hjust = "inward",
    show.legend = FALSE) +
  labs(x = "Number of predictors", y = "Mean/sd value", 
       title = "Plot of the training and test cv error") +
  scale_color_discrete(name = "Legend", labels = c("Train mean", "Train sd", "Test mean", "Test sd"))