如何基于其他列按列获取新列中的值

Question

I am trying to get column desired_output which consists of values based on value column group by grp_1 & grp_2

ie if the values in value column having unique values then values should be NA 's

if values repeats more than any value then entire group will be that repeated value

if values repeats equal times then entire group will be that MAX number value

grp_1 = c("A","A","A","A","A","A","A","A","A","A","A","A","A","A","A","A","A","A","A","A","A")
 grp_2 = c("a","a","a","a","a","b","b","b","b","c","c","c","c","d","d","d","d","e","e","e","e")
 value =c(1,2,3,3,4,1,2,3,4,1,1,2,2,1,2,4,4,1,3,3,3)
desired_output =c(3,3,3,3,3,NA,NA,NA,NA,2,2,2,2,4,4,4,4,3,3,3,3) 

 df = data.frame(grp_1,grp_2,value,desired_output)

I have been struck after getting repeated values count

func <- function(x) { 
  unlist(lapply(rle(x)$lengths, seq_len))

}  

df <- group_by(df,grp_1,grp_2)
df_1 <- mutate(df, common=as.numeric(func(value)) ) 

Thanks in advance

Answer 1

Hope this helps!

library(dplyr)
library(modeest)
final_df <- df %>%
  group_by(grp_1,grp_2) %>%
  mutate(desired_output = ifelse(n()==length(unique(value)),
                                 NA,
                                 ifelse(length(unique(table(value)))==1,
                                        max(value),
                                        mlv(value, method='mfv')[['M']]))) %>%
  data.frame()
final_df

Output is:

   grp_1 grp_2 value desired_output
1      A     a     1              3
2      A     a     2              3
3      A     a     3              3
4      A     a     3              3
5      A     a     4              3
6      A     b     1             NA
7      A     b     2             NA
8      A     b     3             NA
9      A     b     4             NA
10     A     c     1              2
11     A     c     1              2
12     A     c     2              2
13     A     c     2              2
14     A     d     1              4
15     A     d     2              4
16     A     d     4              4
17     A     d     4              4
18     A     e     1              3
19     A     e     3              3
20     A     e     3              3
21     A     e     3              3

#sample data
structure(list(grp_1 = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = "A", class = "factor"), 
    grp_2 = structure(c(1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 
    3L, 3L, 3L, 4L, 4L, 4L, 4L, 5L, 5L, 5L, 5L), .Label = c("a", 
    "b", "c", "d", "e"), class = "factor"), value = c(1, 2, 3, 
    3, 4, 1, 2, 3, 4, 1, 1, 2, 2, 1, 2, 4, 4, 1, 3, 3, 3)), .Names = c("grp_1", 
"grp_2", "value"), row.names = c(NA, -21L), class = "data.frame")

Answer 2

In case someone likes data.table

data.table::setDT(df)

df[,desired_outcome:= max(value[duplicated(value)]), by=c("grp_1","grp_2")
  ][is.infinite(desired_outcome),desired_outcome:=NA]