如何按rep分组压缩数据：保留连续重复Event的首个记录

原始数据 #

df <- data.frame(rep = c(1,1,1,1,1,1,2,2,2,2,2,2),
                 time = c(0.0000,0.0002,0.0003,0.0015,0.0450,0.07800,0.0000,0.0002,0.0003,0.0015,0.0450,0.07800),
                 Event = c("A", "A", "A", "B", "C", "C","A", "B", "B", "B", "C", "C"))

需求 #

按rep分组压缩数据，每组内仅保留连续相同Event序列的第一条记录，直到出现不同Event，目标结果如下：

df2 <- data.frame(rep = c(1,1,1,2,2,2),
                 time = c(0.0000,0.0015,0.0450,0.0000,0.0002,0.0450),
                 Event = c("A","B", "C", "A", "B", "C"))

尝试的代码（存在问题） #

new_df <- df %>%
  group_by(rep) %>%
  slice_head(rle(Event))

注：slice_head()不支持直接传入rle()结果作为参数，这种写法无法实现需求，会导致错误或不符合预期的数据删除。

正确实现方法 #

方法1：结合dplyr与rle #

利用rle()识别连续相同的Event序列，提取每组内每个序列的首行：

library(dplyr)

new_df <- df %>%
  group_by(rep) %>%
  mutate(
    # 生成连续相同Event的分组标记
    group_id = rep(seq_along(rle(Event)$values), rle(Event)$lengths)
  ) %>%
  group_by(rep, group_id) %>%
  slice_head(n = 1) %>% # 取每个连续序列的首条记录
  ungroup() %>%
  select(-group_id) # 移除临时标记列

方法2：用lag()判断Event是否变化 #

通过比较当前行与前一行的Event是否不同，筛选出每组内的首行和Event变化的行：

library(dplyr)

new_df <- df %>%
  group_by(rep) %>%
  filter(
    # 保留每组的第一行，或当前Event与前一行不同的行
    row_number() == 1 | Event != lag(Event)
  ) %>%
  ungroup()

两种方法均可得到目标结果，其中方法2更简洁直观，无需额外生成分组标记。

内容的提问来源于stack exchange，提问作者Werrby