R language de duplication operation unique duplicate filter

For example, there is a data frame , According to the first column ID duplicate removal , Delete all duplicate rows , Do you use unique and duplicate function , It could be wrong , Here's a summary .

Analog data

set.seed(123)
dat = data.frame(ID = c(1:10,9,4,4,9,9,2),y = rnorm(16))
dat

demand ：
hold ID Remove all duplicate lines .

error 1： use unique function

unique function , Will remove the duplicate ID, Keep non duplicate ID, utilize 1,2,3,1 in 1 It's repetitive , use unique And then there is ：1,2,3, instead of 2,3.

therefore , The following steps are wrong .

uid = unique(dat$ID)
dat[dat$ID %in% uid,]

error 2：duplicate function

duplicate Returns the TRUE and FALSE state , Returns a unique value , Instead of removing all duplicates ID Value . similar unique, It's not what we want .

dat[!duplicated(dat$ID),]

correct 1： use filter function

First determine the number of occurrences , extract ID, And then use filter To extract .

uid = dat %>% count(ID) %>% filter(n ==1) %>% select(ID)
uid
dat[dat$ID %in% uid$ID,]

correct 2： use %in%

First use duplicate Print out duplicate ID, And then use filter Just exclude .

uid2  = dat$ID[duplicated(dat$ID)]
uid2

dat %>% filter(!ID %in% uid2)

Complete test code ：

set.seed(123)
dat = data.frame(ID = c(1:10,9,4,4,9,9,2),y = rnorm(16))
dat

#  Wrong way 
dat[!duplicated(dat$ID),]


#  The right way 1
uid = unique(dat$ID)
dat[dat$ID %in% uid,]

uid = dat %>% count(ID) %>% filter(n ==1) %>% select(ID)
uid
dat %>% filter(ID %in% uid$ID)




#  The right way 2
uid2  = dat$ID[duplicated(dat$ID)]
uid2

dat %>% filter(!ID %in% uid2)