独立于顺序汇总 R 数据框中的列，(df$A,df$B) = (df$B,df$A)

Question

我有以下数据框：


    
      destiny origin Count
    1 KJFK    SBBR       4
    2 KJFK    SAEZ    4683
    3 SBGL    KJFK       2
    4 SBBR    KJFK       2
    5 KJFK    SBGL    4987
    6 KJFK    SBGR   12911
    ...

由于我对路线很感兴趣，对我来说，KJFK -> SBBR 与 SBBR -> KJFK 相同。所以我想总结一下他们的数量，如下面的table


    
      destiny origin Count
    1 KJFK    SBBR       6
    2 KJFK    SAEZ    4683
    3 SBGL    KJFK    4989
    4 KJFK    SBGR   12911
    ...

我不想使用大的 for 循环来评估所有值

Answer 1

这个怎么样？

library(tidyverse)
df %>%
    mutate_if(is.factor, as.character) %>%
    rowwise() %>%
    mutate(grp = paste0(sort(c(destiny, origin)), collapse = "_")) %>%
    ungroup() %>%
    group_by(grp) %>%
    summarise(Count = sum(Count)) %>%
    separate(grp, into = c("destiny", "origin"))
#        # A tibble: 4 x 3
#  destiny origin Count
#  <chr>   <chr>  <int>
#1 KJFK    SAEZ    4683
#2 KJFK    SBBR       6
#3 KJFK    SBGL    4989
#4 KJFK    SBGR   12911

请注意，由于您不关心destiny、origin的顺序，所以这里我们按字母顺序排列。因此，在您上面给出的示例中，KJFK -> SBBR 和 SBBR -> KJFK 将变为 destiny = KJFK, origin = SBBR.

示例数据

df <- read.table(text =
    "  destiny origin Count
    1 KJFK    SBBR       4
    2 KJFK    SAEZ    4683
    3 SBGL    KJFK       2
    4 SBBR    KJFK       2
    5 KJFK    SBGL    4987
    6 KJFK    SBGR   12911", header =T)

Answer 2

这里有一个选项pmin/pmax

library(tidyverse)
df1 %>%       
  group_by(destinyN = pmin(destiny, origin), originN = pmax(destiny, origin)) %>% 
  summarise(destiny = first(destiny), 
            origin = first(origin), 
            Count = sum(Count)) %>%
  ungroup %>%
  select(-destinyN, -originN)
# A tibble: 4 x 3
#  destiny origin Count
#  <chr>   <chr>  <int>
#1 KJFK    SAEZ    4683
#2 KJFK    SBBR       6
#3 SBGL    KJFK    4989
#4 KJFK    SBGR   12911

数据

df1 <- structure(list(destiny = c("KJFK", "KJFK", "SBGL", "SBBR", "KJFK", 
"KJFK"), origin = c("SBBR", "SAEZ", "KJFK", "KJFK", "SBGL", "SBGR"
), Count = c(4L, 4683L, 2L, 2L, 4987L, 12911L)), .Names = c("destiny", 
"origin", "Count"), row.names = c("1", "2", "3", "4", "5", "6"
), class = "data.frame")

独立于顺序汇总 R 数据框中的列，(df$A,df$B) = (df$B,df$A)

Summarize columns in R Data Frame independent of the order, (df$A,df$B) = (df$B,df$A)

merge

r

summarization

dataframe

示例数据

数据