计算多个列中重复值的出现次数
Count occurrences of value in multiple columns with duplicates
我的问题非常类似于:
但是,那里提出的解决方案对我不起作用,因为在同一行中,该值可能出现两次,但我只想计算它出现的行数。我已经想出了一个解决方案,但它似乎太长了:
> toy_data = data.table(from=c("A","A","A","C","E","E"), to=c("B","C","A","D","F","E"))
> toy_data
from to
1: A B
2: A C
3: A A
4: C D
5: E F
6: E E
> #get a table with intra-link count
> A = data.table(table(unlist(toy_data[from==to,from ])))
> A
V1 N
1: A 1
2: E 1
A #get a table with total count
> B = data.table(table(unlist(toy_data[,c(from,to)])))
> B
V1 N
1: A 4
2: B 1
3: C 2
4: D 1
5: E 3
6: F 1
>
> # concatenate changing sign
> table = rbind(B,A[,.(V1,-N)],use.names=FALSE)
> # groupby and subtract
> table[,sum(N),by=V1]
V1 V1
1: A 3
2: B 1
3: C 2
4: D 1
5: E 2
6: F 1
是否有一些函数可以在更少的行中完成这项工作?我想在 python 中我会连接 match() 和 match(),但找不到正确的语法
编辑:我知道这行得通 A=length(toy_data[from=="A"|to=="A",from])
但我想避免各种 "A","B"...
之间的循环(而且我不知道如何以这种方式格式化输出)
您可以试试下面的代码
> toy_data[, to := replace(to, from == to, NA)][, data.frame(table(unlist(.SD)))]
Var1 Freq
1 A 3
2 B 1
3 C 2
4 D 1
5 E 2
6 F 1
或
toy_data %>%
mutate(to = replace(to, from == to, NA)) %>%
unlist() %>%
table() %>%
as.data.frame()
这给出了
. Freq
1 A 3
2 B 1
3 C 2
4 D 1
5 E 2
6 F 1
使用data.table
library(data.table)
toy_data[from == to, to := NA][, .(to = na.omit(c(from, to)))][, .N, to]
按照 akrun 的建议使用 to:=NA,可以将结果包装在 table(unlist())
中并转换为 data.table
data.table(table(unlist(toy_data[from==to, to:=NA, from])))
您可以只对 to
向量进行子集化:
data.table(table(unlist(toy_data[,c(from,to[to!=from])])))
V1 N
1: A 3
2: B 1
3: C 2
4: D 1
5: E 2
6: F 1
我的问题非常类似于:
但是,那里提出的解决方案对我不起作用,因为在同一行中,该值可能出现两次,但我只想计算它出现的行数。我已经想出了一个解决方案,但它似乎太长了:
> toy_data = data.table(from=c("A","A","A","C","E","E"), to=c("B","C","A","D","F","E"))
> toy_data
from to
1: A B
2: A C
3: A A
4: C D
5: E F
6: E E
> #get a table with intra-link count
> A = data.table(table(unlist(toy_data[from==to,from ])))
> A
V1 N
1: A 1
2: E 1
A #get a table with total count
> B = data.table(table(unlist(toy_data[,c(from,to)])))
> B
V1 N
1: A 4
2: B 1
3: C 2
4: D 1
5: E 3
6: F 1
>
> # concatenate changing sign
> table = rbind(B,A[,.(V1,-N)],use.names=FALSE)
> # groupby and subtract
> table[,sum(N),by=V1]
V1 V1
1: A 3
2: B 1
3: C 2
4: D 1
5: E 2
6: F 1
是否有一些函数可以在更少的行中完成这项工作?我想在 python 中我会连接 match() 和 match(),但找不到正确的语法
编辑:我知道这行得通 A=length(toy_data[from=="A"|to=="A",from])
但我想避免各种 "A","B"...
之间的循环(而且我不知道如何以这种方式格式化输出)
您可以试试下面的代码
> toy_data[, to := replace(to, from == to, NA)][, data.frame(table(unlist(.SD)))]
Var1 Freq
1 A 3
2 B 1
3 C 2
4 D 1
5 E 2
6 F 1
或
toy_data %>%
mutate(to = replace(to, from == to, NA)) %>%
unlist() %>%
table() %>%
as.data.frame()
这给出了
. Freq
1 A 3
2 B 1
3 C 2
4 D 1
5 E 2
6 F 1
使用data.table
library(data.table)
toy_data[from == to, to := NA][, .(to = na.omit(c(from, to)))][, .N, to]
按照 akrun 的建议使用 to:=NA,可以将结果包装在 table(unlist())
中并转换为 data.table
data.table(table(unlist(toy_data[from==to, to:=NA, from])))
您可以只对 to
向量进行子集化:
data.table(table(unlist(toy_data[,c(from,to[to!=from])])))
V1 N
1: A 3
2: B 1
3: C 2
4: D 1
5: E 2
6: F 1