统一使用字符串中的单位数和双位数

Homogenize use of single and double digit numbers in string

我有一个非常大的 data.table,其中(大量)项目由包括文本和数字的字符串定义。

library(data.table)    
dd <- data.table(x = c("A4","A4","A4","A14","A14","A14","B4","B4","B4"),y = c("A4","A14","B4","A4","A14","B4","A4","A14","B4"), z = c(1,2,3,4,5,6,7,8,9))

x   y   z
A4  A4  1
A4  A14 2
A4  B4  3
A14 A4  4
A14 A14 5
A14 B4  6
B4  A4  7
B4  A14 8
B4  B4  9

数字可以是个位数或双位数,因此 R 将始终根据数字中的第一个数字(A14 在 A4 之前)对它们进行排序。混合排序可以处理这个。但是,当我将长数据重塑为宽数据时

wide <- dcast(dd, x ~ y, value.var = "z")

R 正在根据基本排序规则再次应用排序。

x    A14  A4  B4
A14  5    4   6
A4   2    1   3
B4   8    7   9

但是我需要原始顺序来进行后续矩阵计算。有什么有效的方法可以将字符串+单个数字重命名为字符串+双位数(A4-> A04)或我错过的另一种方法吗?

您可以使用 sprintf() 为数字预填充 0

sprintf("%s%02.0d", "A",  1:20)
# [1] "A01" "A02" "A03" "A04" "A05" "A06" "A07" "A08" "A09" "A10" "A11" "A12" "A13" "A14" "A15" "A16" "A17" "A18" "A19" "A20"

您可以使用

0 添加到您的数据中
dd[nchar(x) == 2, x := paste0(substr(x, 1, 1), 0, substr(x, 2, 2))]
dd[nchar(y) == 2, y := paste0(substr(y, 1, 1), 0, substr(y, 2, 2))]

#      x   y z
# 1: A04 A04 1
# 2: A04 A14 2
# 3: A04 B04 3
# 4: A14 A04 4
# 5: A14 A14 5
# 6: A14 B04 6
# 7: B04 A04 7
# 8: B04 A14 8
# 9: B04 B04 9

或者,如果您需要申请更多栏目:

to.change <- c('x', 'y')

dd[, (to.change) := lapply(.SD, function(x) ifelse(nchar(x) > 2, x
                                                   , paste0(substr(x, 1, 1), 0, substr(x, 2, 2))))
   , .SDcols = to.change]

此解决方案不需要额外的零。

# Data frame
df <- data.frame(x = c("A4","A4","A4","A14","A14","A14","B4","B4","B4"),
                 y = c("A4","A14","B4","A4","A14","B4","A4","A14","B4"), 
                 z = c(1,2,3,4,5,6,7,8,9),
                 stringsAsFactors = FALSE)

# Reorder columns and rows using `mixedsort`. 
wide <- dcast(df, x ~ y,value.var   = "z") %>% 
  select(x, mixedsort(unique(df$x))) %>% 
  slice(match(x, mixedsort(unique(df$x))))

给予,

#     x A4 A14 B4
# 1  A4  1   2  3
# 2 A14  4   5  6
# 3  B4  7   8  9

另一个可能是最简单的选项是使用 gtools-package 中的 mixedorder

wide <- dcast(dd, x ~ y, value.var = "z")[gtools::mixedorder(x)]

给出:

> wide
     x A14 A4 B4
1:  A4   2  1  3
2: A14   5  4  6
3:  B4   8  7  9

如果您还想以相同的方式设置列顺序,您还可以使用 setcolorder:

setcolorder(wide, c(1, gtools::mixedorder(names(wide)[-1]) + 1))

然后给出:

> wide
     x A4 A14 B4
1:  A4  1   2  3
2: A14  4   5  6
3:  B4  7   8  9

您可能需要考虑通过因子直接在数据中实现此顺序,这样您以后就不必通过数据整理来修复它。

如果您已经将这些唯一值排序到某处,您将不需要 mixedorder 而不是 mixedsort,然后将它们转换为因子。

否则您可以取回订单:

library(gtools)
dd[,1:2] <- lapply(dd[,1:2],function(x) factor(x, mixedsort(unique(x))))

并正常进行:

dcast(dd, x ~ y, value.var = "z")
#      x A4 A14 B4
# 1:  A4  1   2  3
# 2: A14  4   5  6
# 3:  B4  7   8  9