我正在嘗試pivot_longer使用多個變數集，但無法從示例中正確獲取語法。
我的虛擬資料集是：
library(dplyr)
library(tidyr)
ID = c("id-1", "id-2", "id-3")
State = c("MD", "MD", "VA")
Time1Day= c( 1, 12, 30)
Time1Month = c( 1, 4, 5)
Time2Day = c( 9, 21, 13)
Time2Month = c( 12, 4, 5)
Time3Day = c( 7, 14, NA)
Time3Month = c( 1, 2, NA)
df <-data.frame(ID, State, Time1Day, Time1Month, Time2Day, Time2Month, Time3Day, Time3Month)
我想要的結果是：
ID State Time Day Month
1 id-1 MD Time1 1 1
2 id-1 MD Time2 9 12
3 id-1 MD Time3 7 1
4 id-2 MD Time1 12 4
5 id-2 MD Time2 21 4
6 id-2 MD Time3 14 2
7 id-3 VA Time1 30 5
8 id-3 VA Time2 13 5
我查看了here和here以嘗試正確使用語法，并嘗試了以下兩種解決方案，但我無法正常作業：
df.long <- df %>%
pivot_longer(cols = starts_with("Time"), names_to = c("Day", "Month"), names_sep="(?=[0-9])"), values_to = "Time", values_drop_na = TRUE)
df.long <- df %>%
pivot_longer(cols = ends_with("Day"), names_to = c("Time"), values_to = "Days", values_drop_na = TRUE) %>%
pivot_longer(cols = ends_with("Month"), names_to = c("Time"), values_to = "Months", values_drop_na = TRUE)
任何關于我缺少什么以及如何解決它的建議將不勝感激
uj5u.com熱心網友回復：
編輯添加values_drop_na = TRUE感謝 TarJae 的評論。
你可以用
library(dplyr)
library(tidyr)
df %>%
pivot_longer(-c(ID, State),
names_to = c("Time", ".value"),
names_pattern = "(Time\\d)(.*)",
values_drop_na = TRUE)
這回傳
# A tibble: 9 x 5
ID State Time Day Month
<chr> <chr> <chr> <dbl> <dbl>
1 id-1 MD Time1 1 1
2 id-1 MD Time2 9 12
3 id-1 MD Time3 7 1
4 id-2 MD Time1 12 4
5 id-2 MD Time2 21 4
6 id-2 MD Time3 14 2
7 id-3 VA Time1 30 5
8 id-3 VA Time2 13 5
uj5u.com熱心網友回復：
一種data.table方法
library(data.table)
# melt to long
DT <- melt(setDT(df), id.vars = c("ID", "State"), variable.factor = FALSE, na.rm = TRUE)
# split variable string
DT[, c("Time", "part2") := tstrsplit(variable, "(?<=[0-9])", perl=TRUE)]
# recast to wide
dcast(DT, ID State Time ~ part2, value.var = "value", drop = TRUE)
# ID State Time Day Month
# 1: id-1 MD Time1 1 1
# 2: id-1 MD Time2 9 12
# 3: id-1 MD Time3 7 1
# 4: id-2 MD Time1 12 4
# 5: id-2 MD Time2 21 4
# 6: id-2 MD Time3 14 2
# 7: id-3 VA Time1 30 5
# 8: id-3 VA Time2 13 5
轉載請註明出處，本文鏈接：https://www.uj5u.com/ruanti/343663.html