CODEWITHSUNDEEP
rif-statementapply
user908645
user908645

Reputation: 33

condition statement in apply function not work correctly in R

I had a dataframe, whose ID column had many duplicated names. So I used table() function to get the frequency of IDs. like this:

library(dplyr)
id <- runif(1000,1000,3000) %>% round() %>% as.character()
freq <- rep(1:50,20)
data <- data.frame(id,freq)
GetID <- function(a){
  if (a[2]==1) newid <- a[1] else newid <- paste(a[1],1:a[2],sep = "-");
  return(newid)}
idlist <- data %>% apply(., 1, GetID)
idlist2 <- unlist(idlist) %>% as.data.frame()

I wanted to get a new ID vector. If the freq equals 1, the new ID equals the old one. If the freq is larger than 1, the new ID is the old id combined with its order. However, it seems the if statement didn't work correctly. All of new id had order number.

Upvotes: 0

Views: 284

Answers (2)

mohsen0965
mohsen0965

Reputation: 77

do you have to use a function? if not:

id <- runif(1000,1000,3000)
freq <- rep(1:50,20)
num <- 1:length(id)
data <- data.frame(num,id,freq)
data2 <- data %>% filter(freq == 1) %>% mutate(newid = id)
data3 <- data %>% filter(freq != 1) %>% mutate(newid = paste(id,freq,sep = "-"))
result <- rbind(data2,data3) %>% arrange(num)

Upvotes: 1

Ronak Shah
Ronak Shah

Reputation: 388982

You can group_by id and if number of rows is greater than 1 then paste row_number() with id or just use id.

library(dplyr)

data %>%
  group_by(id) %>%
  mutate(newID = if(n() > 1) paste(id, row_number(), sep = '-') 
                 else as.character(id)) %>%
  arrange(id)


#    id     freq newID 
#   <chr> <int> <chr> 
# 1 1002     49 1002-1
# 2 1002     31 1002-2
# 3 1003     26 1003  
# 4 1005     11 1005-1
# 5 1005     28 1005-2
# 6 1007     37 1007  
# 7 1013     33 1013  
# 8 1016      7 1016  
# 9 1020     11 1020  
#10 1024     28 1024  
# … with 990 more rows

Upvotes: 0

Related Questions