问题解释
给定两个可变长度的向量，并且向量之间或向量内没有重复，如何有效地按升序和交替顺序组合向量？
下面是一个示例：
a <- c(98, 101, 104, 136, 154, 193)
b <- c(31, 37, 41, 44, 48, 55, 80, 118, 179)
预期输出为
c(31, 98, 118, 136, 179, 193)
# b, a, b, a, b, a
你可以看到我们从 31 开始a，向量之间的最小值。 其次是b的 98 .然后从a开始，下一个大于98的数是118。依此类推，导致：
A = 98, 136, 193
B = 31, 118, 179
我的尝试：
x <- c(min(a,b))
lastwas <- startedwithA <- ifelse(x %in% a, 1, 2)
for(i in 1:(length(a)+length(b))){
if(lastwas == 2){
x <- c(x, a[which(a > x[i])[1]])
lastwas <- 1
} else if(lastwas == 1){
x <- c(x, b[which(b > x[i])[1]])
lastwas <- 2
}
}
(x <- x[!is.na(x)])
# [1] 31 98 118 136 179 193
if(startedwithA == 1){
evenodd <- c(T,F)
} else {
evenodd <- c(F,T)
}
(A = x[evenodd])
# [1] 98 136 193
(B = x[!evenodd])
# [1] 31 118 179
有没有更好的方法来解决这个问题？
你可以像这样使用data.frame、order和diff：
a = c(98, 101, 104, 136, 154, 193)
b = c(31, 37, 41, 44, 48, 55, 80, 118, 179)
# Create a data frame for each vector with a different number in column 'set'
dfa = data.frame(val = a, set = 1)
dfb = data.frame(val = b, set = 2)
# Bind both together and order them by val
df = rbind(dfa,dfb)
df = df[order(df$val),]
# Only keep those that have a diff different to 0
keep = c(1, diff(df$set)) != 0
result = df[keep,]
结果：
val set
7 31 2
1 98 1
14 118 2
4 136 1
15 179 2
6 193 1
您可以根据向量设置名称前缀，
v <- sort(c(setNames(a, paste0("a", a)), setNames(b, paste0("b", b))))
并使用分解初始字符的diff。
res <- v[!!c(1, diff(as.numeric(as.factor(substr(names(v), 1, 1)))))]
res
# b31 a98 b118 a136 b179 a193
# 31 98 118 136 179 193
然后拆分为数据框：
d <- as.data.frame(split(res, substr(names(res), 1, 1)))
d
# a b
# a98 98 31
# a136 136 118
# a193 193 179
数据
a <- c(98, 101, 104, 136, 154, 193)
b <- c(31, 37, 41, 44, 48, 55, 80, 118, 179)