在 dplyr::summarise 中传递一个连接的字符串作为列名

Question

我正在尝试使用连接的字符串作为列名迭代地执行 dplyr 汇总

Category=c("a","a","b","b","b","c","c","c")
A1=c(1,2,3,4,3,2,1,2)
A2=c(10,11,12,13,14,15,16,17)
tt=cbind(Category,A1,A2)
tdat=data.frame(tt)
colnames(tdat)=c("Category","M1","M2")
ll=matrix(1:2,nrow=2)
for(i in 1:nrow(ll)) {
  Aone=tdat %>% group_by(Category) %>%
    summarize(Msum=sum(paste("M",i,sep="")))
}

我最后报了下面的错误

x invalid 'type' (character) of argument
ℹ Input Msum is sum(paste("M", i, sep = "")).
ℹ The error occurred in group 1: Category = "A".
Run rlang::last_error() to see where the error occurred.```


The goal is to iteratively get arithmentic functions within summarize function in dplyr. But this concatenated string is not recognized as column name.

Answer 1

如果我们想将字符串作为列名传递，则转换为 symbol 并计算 (!!)

library(dplyr)
Aone <- vector('list', nrow(ll))
for(i in seq_len(nrow(ll))) {
      Aone[[i]] <- tdat %>%
                    group_by(Category) %>%
                    summarize(Msum = sum(!! rlang::sym(paste("M", i, sep=""))))
    }

或者假设列名是 'M-1'、'M-2' 等，它应该也能工作

Aone <- vector('list', 2)
for(i in seq_along(Aone)) {
   Aone[[i]] <- tdat %>%
        group_by(Category) %>% 
       summarise(Msum = sum(!! rlang::sym(paste("M-", i, sep=""))), 
         .groups = 'drop')
  }

注意：ll 在原 post 中并不清楚。在这里，我们创建一个 list，其中 length 等于 'M-' 列的数量，并通过遍历 [=16] 的序列将输出分配回 list 元素=]

数据

tdat <- data.frame(Category, M1, M2)


tdat <- structure(list(Category = c("A", "A", "A", "A", "B", "B", "B", 
"B"), `M-1` = c(1, 2, 3, 4, 3, 2, 1, 2), `M-2` = c(10, 11, 12, 
13, 14, 15, 16, 17)), class = "data.frame", row.names = c(NA, 
-8L))

在 dplyr::summarise 中传递一个连接的字符串作为列名

Pass a concatenated string as column name in dplyr::summarise

r

concatenation

paste

dplyr

summarize

数据