Duplicated 与unique

>DT <- data.table(A = rep(1:3, each=4), B = rep(1:4, each=3),
+                  C = rep(1:2, 6), key = "A,B")

> DT
    A B C
 1: 1 1 1
 2: 1 1 2
 3: 1 1 1
 4: 1 2 2
 5: 2 2 1
 6: 2 2 2
 7: 2 3 1
 8: 2 3 2
 9: 3 3 1
10: 3 4 2
11: 3 4 1
12: 3 4 2

 

> duplicated(DT)      #R中默认的是fromLast=FALSE,即若样本点重复出现，则取首次出现的；否则去最后一次出现的
 [1] FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE

 

> unique(DT) #把重复的两行去掉，即上面的TRUE对应的行
    A B C
 1: 1 1 1
 2: 1 1 2
 3: 1 2 2
 4: 2 2 1
 5: 2 2 2
 6: 2 3 1
 7: 2 3 2
 8: 3 3 1
 9: 3 4 2
10: 3 4 1

 

> duplicated(DT, by="B") #只看第“B”列
 [1] FALSE  TRUE  TRUE FALSE  TRUE  TRUE FALSE  TRUE  TRUE FALSE  TRUE  TRUE
> unique(DT, by="B")#按照B列的结果返回数据，包含三列的数据
   A B C
1: 1 1 1
2: 1 2 2
3: 2 3 1
4: 3 4 2
> duplicated(DT, by=c("A", "C"))#按照A\C两列的数据返回结果，原理同上
 [1] FALSE FALSE  TRUE  TRUE FALSE FALSE  TRUE  TRUE FALSE FALSE  TRUE  TRUE
> unique(DT, by=c("A", "C"))
   A B C
1: 1 1 1
2: 1 1 2
3: 2 2 1
4: 2 2 2
5: 3 3 1
6: 3 4 2

 

#data.table中duplicated与unique的用法

 

> DT = data.table(a=c(2L,1L,2L), b=c(1L,2L,1L))   # no key  

> DT
   a b
1: 2 1
2: 1 2
3: 2 1

> unique(DT)                   # rows 1 and 2 (row 3 is a duplicate of row 1)
   a b
1: 2 1
2: 1 2
> DT = data.table(a=c(3.142, 4.2, 4.2, 3.142, 1.223, 1.223), b=rep(1,6))

> DT
       a b
1: 3.142 1
2: 4.200 1
3: 4.200 1
4: 3.142 1
5: 1.223 1
6: 1.223 1

> unique(DT)                   # rows 1,2 and 5
       a b
1: 3.142 1
2: 4.200 1
3: 1.223 1
> DT = data.table(a=tan(pi*(1/4 + 1:10)), b=rep(1,10))   # example from ?all.equal

> DT
    a b
 1: 1 1
 2: 1 1
 3: 1 1
 4: 1 1
 5: 1 1
 6: 1 1
 7: 1 1
 8: 1 1
 9: 1 1
10: 1 1

> length(unique(DT$a))         # 10 strictly unique floating point values
[1] 10

 

> all.equal(DT$a,rep(1,10))    # TRUE, all within tolerance of 1.0  #
[1] TRUE
> DT[,which.min(a)]            # row 10, the strictly smallest floating point value
[1] 10
> identical(unique(DT),DT[1])  # TRUE, stable within tolerance
[1] FALSE
> identical(unique(DT),DT[10]) # FALSE
[1] FALSE

 

 

 

 

 

 

 

 

 

 

 

 

duplicated(DT)

unique(DT)

 

duplicated(DT, by="B")

unique(DT, by="B")

 

duplicated(DT, by=c("A", "C"))

unique(DT, by=c("A", "C"))

 

#data.table 中的用法 一定要弄清楚

DT = data.table(a=c(2L,1L,2L), b=c(1L,2L,1L))   # no key

unique(DT)                   # rows 1 and 2 (row 3 is a duplicate of row 1)

 

DT = data.table(a=c(3.142, 4.2, 4.2, 3.142, 1.223, 1.223), b=rep(1,6))

unique(DT)                   # rows 1,2 and 5

 

DT = data.table(a=tan(pi*(1/4 + 1:10)), b=rep(1,10))   # example from ?all.equal

length(unique(DT$a))         # 10 strictly unique floating point values

#a中都是浮点型数据，所以每个值都不一样；b中都是整型数据，所以都是严格相等的

 

 

all.equal(DT$a,rep(1,10))    # TRUE, all within tolerance of 1.0  #

#

DT[,which.min(a)]            # row 10, the strictly smallest floating point value

#第10行是浮点型数据最小的

identical(unique(DT),DT[1])  # TRUE, stable within tolerance

identical(unique(DT),DT[10]) # FALSE

 

# fromLast=TRUE

DT <- data.table(A = rep(1:3, each=4), B = rep(1:4, each=3),

                 C = rep(1:2, 6), key = "A,B")

duplicated(DT, by="B", fromLast=TRUE)

unique(DT, by="B", fromLast=TRUE)

 

# anyDuplicated

anyDuplicated(DT, by=c("A", "B"))    # 3L

any(duplicated(DT, by=c("A", "B")))  # TRUE

 

# uniqueN, unique rows on key columns

uniqueN(DT, by = key(DT))

# uniqueN, unique rows on all columns

uniqueN(DT)

# uniqueN while grouped by "A"

DT[, .(uN=uniqueN(.SD)), by=A]

 

# uniqueN's na.rm=TRUE

x = sample(c(NA, NaN, runif(3)), 10, TRUE)

uniqueN(x, na.rm = FALSE) # 5, default

uniqueN(x, na.rm=TRUE) # 3