仅在ggplot中显示变量点标签的第一个实例

Question

我有一个像这样的数据框

GROUP <- c("PRE_ANALYSIS","PRE_ANALYSIS","BEFORE","AFTER","AFTER","AFTER","AFTER","AFTER","AFTER","AFTER","AFTER")
PDATETIME <- c("2016-11-23 06:08:39","2017-02-25 15:31:50","2017-04-06 17:11:57","2017-04-15 01:23:51 CDT",
               "2017-04-16 15:54:35","2017-04-16 23:48:14","2017-04-18 02:57:41","2017-04-20 05:12:33 CDT",
               "2017-04-20 14:33:20","2017-04-22 12:33:44","2017-04-23 10:50:49")
DELTA <- c(766.3770,863.8409,893.1615,656.1001,832.9232,912.3867,934.9714,-64.7182,460.3154,949.2538,697.3750)

df <- data.frame(GROUP,PDATETIME,DELTA) 
df$PDATETIME <- as.POSIXct(df$PDATETIME,format="%Y-%m-%d %H:%M:%S")

我正在尝试创建一个新列＆＃34; IsChangePoint＆＃34;其值为

"CP Detected" for the first changepoint, the first index of df$GROUP = "AFTER"

我试图这样做

df$IsChangepoint <- ifelse(df$GROUP == "AFTER", "CP Detected", "No")

然后得到这个情节

library(ggplot2)
library(ggrepel)
ggplot(data = df, aes(PDATETIME,DELTA ))+ 
  ggtitle("Changepoint Analysis") + 
  theme(axis.text.x = element_text(angle=90, vjust=1),plot.title = element_text(size = rel(1))) + 
  geom_point(aes(colour=df$GROUP), size = 6) +
  geom_text_repel(aes(PDATETIME, DELTA, label = df$IsChangepoint)) +
  labs(colour="GROUP") + 
  xlab("PDATETIME")+ 
  ylab("DELTA")

这标记变更点之后的所有索引。如何绘制＆＃34; CP DETECTED＆＃34;的第一个实例？

所需的情节：

有人能指出我正确的方向吗？

Answer 1

看起来您希望IsChangepoint采用3个值中的1个：

1. GROUP！=＆＃34; After＆＃34;：＆＃34; No＆＃34; （编辑基于OP的评论：这也应该是空白的）
2. GROUP ==＆＃34; AFTER＆＃34; ＆安培;该组中的第1行：＆＃34; CP检测到＆＃34;
3. GROUP ==＆＃34; AFTER＆＃34; ＆安培;不是这一组中的第一行：＆＃34;＆＃34; （空白）

如果是这种情况，以下内容将在数据集中为您提供：

df <- df %>% 
  arrange(PDATETIME) %>%
  group_by(GROUP) %>% mutate(IsChangepoint = row_number() == 1) %>% ungroup() %>% 
  mutate(IsChangepoint = ifelse(GROUP == "AFTER" & IsChangepoint, "CP Detected", ""))

你原来的ggplot代码对我有用，虽然我认为确实没有必要在df$电话中放geom_text_repel() ...

ggplot(data = df, aes(PDATETIME,DELTA ))+ 
  ggtitle("Changepoint Analysis") + 
  theme(axis.text.x = element_text(angle=90, vjust=1),plot.title = element_text(size = rel(1))) + 
  geom_point(aes(colour=df$GROUP), size = 6) +
  geom_text_repel(aes(PDATETIME, DELTA, label = IsChangepoint)) +
  labs(colour="GROUP") + 
  xlab("PDATETIME")+ 
  ylab("DELTA")