r如果我有两个内容相同的独立组，如何删除其中一个

Question

我的代码正在处理的实际数据帧比这个大得多，它需要能够处理不同的数据帧。下面的例子说明了具有相同内容的组的问题，以及如何只保留其中一个组。

考虑一下我有不同内容的组的情况。

Group   Contents
GroupA  Marble
GroupB  Marble
GroupB  Granite
GroupC  Marble
GroupD  Granite
GroupD  Glass
GroupD  Marble

在上面的例子中，GroupA和GroupC都只包含大理石，所以我想删除其中一个组。我想要的输出：

Group   Contents
GroupA  Marble
GroupB  Marble
GroupB  Granite
GroupD  Granite
GroupD  Glass
GroupD  Marble

可再现数据：

structure(list(Group = c("GroupA", "GroupB", "GroupB", "GroupC", 
"GroupD", "GroupD", "GroupD"), Contents = c("Marble", "Marble", 
"Granite", "Marble", "Granite", "Glass", "Marble")), class = "data.frame", row.names = c(NA, 
-7L), spec = structure(list(cols = list(Group = structure(list(), class = c("collector_character", 
"collector")), Contents = structure(list(), class = c("collector_character", 
"collector"))), default = structure(list(), class = c("collector_guess", 
"collector")), skip = 1), class = "col_spec")) 

Answer 1

您可以尝试：

idx <- !duplicated(with(df, cbind(Contents, ave(Contents, Group, FUN = function(x) toString(sort(x))))))

df[idx, ]
Group Contents
1 GroupA   Marble
2 GroupB   Marble
3 GroupB  Granite
5 GroupD  Granite
6 GroupD    Glass
7 GroupD   Marble

Answer 2

这里有一个通过嵌套aggregate的选项

df[df$Group %in% aggregate(Group~.,aggregate(.~Group,df,toString),head,1)$Group,]

• 输出

Group Contents
1 GroupA   Marble
2 GroupB   Marble
3 GroupB  Granite
5 GroupD  Granite
6 GroupD    Glass
7 GroupD   Marble

Answer 3

dplyr中带有distinct的选项

library(dplyr)
df %>% 
arrange(across(everything())) %>%
group_by(Group) %>%
mutate(new = toString(Contents)) %>%
ungroup %>%
distinct(Contents, new, .keep_all = TRUE) %>%
select(-new)

-输出

# A tibble: 6 x 2
#  Group  Contents
#  <chr>  <chr>   
#1 GroupA Marble  
#2 GroupB Granite 
#3 GroupB Marble  
#4 GroupD Glass   
#5 GroupD Granite 
#6 GroupD Marble  

Answer 4

您可以使用dplyr过滤函数：-

df<-structure(list(Group = c("GroupA", "GroupB", "GroupB", "GroupC", 
"GroupD", "GroupD", "GroupD"), Contents = c("Marble", "Marble", 
"Granite", "Marble", "Granite", "Glass", "Marble")), class = "data.frame", row.names = c(NA, 
                                                                                        -7L), spec = structure(list(cols = list(Group = structure(list(), class = c("collector_character", 
                                                                                                                                                                    "collector")), Contents = structure(list(), class = c("collector_character", 
                                                                                                                                                                                                                          "collector"))), default = structure(list(), class = c("collector_guess", 
                                                                                                                                                                                                                                                                                "collector")), skip = 1), class = "col_spec")) 
df
df%>%
filter(Group!="GroupC")

Group Contents
1 GroupA   Marble
2 GroupB   Marble
3 GroupB  Granite
4 GroupD  Granite
5 GroupD    Glass
6 GroupD   Marble