Creating list of matrices from clusters of vectors with equivalent indices

0
votes

I've got a two different vectors - one with zeros and random real numbers between 0-0.5 (vec1) and another ordered vector (vec2):

vec1 <- c(0.42887017, 0.26703377, 0, 0, 0, 0.33203175, 0.16787991, 0, 0, 0.19483491, 0.41869476, 0.05820833, 0.37449489, 0, 0, 0, 0, 0, 0.44390140, 0.19483491, 0.06736238, 0.31630117, 0, 0, 0, 0, 0, 0, 0.27121130, 0)
vec2 <- c(-0.1, -0.1, -0.1, -0.1, -0.1, 1.2, 1.2, 1.2, 1.2, 1.2, 0.5, 0.5, 0.5, 0.5, 0.5, 2.0, 2.0, 2.0, 2.0, 2.0, -0.6, -0.6, -0.6, -0.6, -0.6, 0.25, 0.25, 0.25, 0.25, 0.25)

For the first vector, vec1, I want create clusters of vectors > 0 and in the second vector, vec2, I want the equivalently positioned indexed elements to be clustered as follows (see bold):

vec1 -> 0.42887017, 0.26703377, 0, 0, 0, 0.33203175, 0.16787991, 0, 0, 0.19483491, 0.41869476, 0.05820833, 0.37449489, 0, 0, 0, 0, 0, 0.44390140, 0.19483491, 0.06736238, 0.31630117, 0, 0, 0, 0, 0, 0, 0.27121130, 0

vec2 -> -0.1, -0.1, -0.1, -0.1, -0.1, 1.2, 1.2, 1.2, 1.2, 1.2, 0.5, 0.5, 0.5, 0.5, 0.5, 2.0, 2.0, 2.0, 2.0, 2.0, -0.6, -0.6, -0.6, -0.6, -0.6, 0.25, 0.25, 0.25, 0.25, 0.25

Preferably the output should be in a lists of matrices with the equivalent indices:

[[1]]
          [,1] [,2]
[1,] 0.4288702 -0.1
[2,] 0.2670338 -0.1

[[2]]
          [,1] [,2]
[1,] 0.3320318  1.2
[2,] 0.1678799  1.2

[[3]]
           [,1] [,2]
[1,] 0.19483491  1.2
[2,] 0.41869476  0.5
[3,] 0.05820833  0.5
[4,] 0.37449489  0.5

[[4]]
           [,1] [,2]
[1,] 0.44390140  2.0
[2,] 0.19483491  2.0
[3,] 0.06736238 -0.6
[4,] 0.31630117 -0.6

[[5]]
          [,1] [,2]
[1,] 0.2712113 0.25

Has anybody got some ideas on how to do this?

Thanks

2
rlistmatrixindexingcluster-computing
what do you mean by equally indexed? By the looks of your example it looks like you want to retrieve the index of elements which are greater than 9 in vec1? – monte
Sorry maybe my wording isn't right. I want to have the elements of vec2 to be clustered in the same position as vec1. Then I want to combine the corresponding clusters into 2-columned matrices. – AlexLee
@Huntmerson please see my solution below and upvote it and accept it if it does what your original question requires. – hello_friend

2 Answers

0
votes

(Current Question) Base R solution:

# Cluster the data into groups, for each series of data above 0:
clustered <- subset(within(data.frame(cbind(vec1, vec2)),
                           {
                             grp <- cumsum(c(TRUE, diff(vec1) == vec1[-1]))
                           }), vec1 > 0)

# Split the dataframe into a list for each group, remove group vector:
setNames(split(within(clustered, rm("grp")), clustered$grp), 
         c(1:length(unique(clustered$grp))))

Current Data:

vec1 <- c(0.42887017, 0.26703377, 0, 0, 0, 0.33203175, 0.16787991, 0, 0, 0, 0.41869476, 0.05820833, 0.37449489, 0, 0, 0, 0, 0, 0.44390140, 0.19483491, 0.06736238, 0.31630117, 0, 0, 0, 0, 0, 0, 0.27121130, 0)
vec2 <- c(-0.1, -0.1, -0.1, -0.1, -0.1, 1.2, 1.2, 1.2, 1.2, 1.2, 0.5, 0.5, 0.5, 0.5, 0.5, 2.0, 2.0, 2.0, 2.0, 2.0, -0.6, -0.6, -0.6, -0.6, -0.6, 0.25, 0.25, 0.25, 0.25, 0.25)

(Original Question) Base R solution:

clustered <- subset(within(data.frame(cbind(vec1, vec2)), 
       {grp <- cumsum(c(TRUE, abs(diff(vec1 > 9))))}), vec1 > 9)

setNames(Map(function(x){within(x, rm("grp"))}, 
    split(clustered, clustered$grp)), c(1:length(unique(clustered$grp))))
-1
votes

I managed to find a way that works myself, maybe a bit complex:

list1 = list()

clust = c()
clust2 = c()

x = 1

for (i in 1:length(vec1)) {
  if (vec1[i] > 0 & i != length(vec1)) {
    clust = c(clust,vec1[i])
    clust2 = c(clust2,vec2[i])
  } else if (vec1[i] == 0 & length(clust) > 0) {
    list1[[x]] <- cbind(clust,clust2)
    
    x = x + 1
    clust = c()
    clust2 = c()
  } else if (i == length(vec1) & vec1[length(vec1)] > 0){
    clust = c(clust ,vec1[i])
    clust2 = c(clust2,vec2[i])
    
    list1[[x]] <- cbind(clust,clust2)
    
  } else {
  }
}

The output is:

> list1
[[1]]
         clust clust2
[1,] 0.4288702   -0.1
[2,] 0.2670338   -0.1

[[2]]
         clust clust2
[1,] 0.3320318    1.2
[2,] 0.1678799    1.2

[[3]]
          clust clust2
[1,] 0.19483491    1.2
[2,] 0.41869476    0.5
[3,] 0.05820833    0.5
[4,] 0.37449489    0.5

[[4]]
          clust clust2
[1,] 0.44390140    2.0
[2,] 0.19483491    2.0
[3,] 0.06736238   -0.6
[4,] 0.31630117   -0.6

[[5]]
         clust clust2
[1,] 0.2712113   0.25