Looping cumulatively including the last variable from a vector-CodePudding

I have a network/graph (test_FINAL), from which I want to remove a certain set of vertices below:

vertex <- c(1, 3, 7, 9, 10)

The idea is to run a clustering analyses after the vertices are removed. For example, in the first iteration, vertex #1 should be removed and then, vertices #1 and #3, and the subsequent #1, #3, #7 and so on until the final iteration where all the vertices are removed and clustering performed.

My code looks as follows:

# Library
library(igraph)

# Create data
set.seed(1)
data <- matrix(sample(0:1, 100, replace=TRUE, prob=c(0.8,0.2)), nc=10)
network <- graph_from_adjacency_matrix(data , mode='undirected', diag=F )

# Default network
par(mar=c(0,0,0,0))
plot(network)

vertex <- c(1, 3, 7, 9, 10)

# Estimating cluster statistics *prior to* removing vertices
pre_cluster <- cluster_fast_greedy(network)
length(pre_cluster); sizes(pre_cluster); modularity(pre_cluster)


# removing vertices
final_graph <- delete_vertices(network, c(vertex))
cluster_graph <- cluster_fast_greedy(final_graph)


# Estimating cluster statistics *after* removing vertices
length(cluster_graph); sizes(cluster_graph); modularity(cluster_graph)

What is the best way to do this in a loop, especially for the final iteration where all the vertices are removed and then the cluster statistics are estimated?

CodePudding user response：

How about the following

vertex <- c(1, 3, 7, 9, 10)

cumulator <- vertex[1]

while (cumulator <= length(vertex)) {
  to_be_deleted <- vertex[c(1:cumulator)]
  cumulator = cumulator   1
  print(to_be_deleted)
}

And in each iteration of the while loop, you could then run your deletion function.

CodePudding user response：

Probably you can start from using Reduce to create a cumulative list of vertices

> Reduce(c, vertex, accumulate = TRUE)
[[1]]
[1] 1

[[2]]
[1] 1 3

[[3]]
[1] 1 3 7

[[4]]
[1] 1 3 7 9

[[5]]
[1]  1  3  7  9 10

and then try

lapply(
    Reduce(c, vertex, accumulate = TRUE),
    function(x) {
        cluster_fast_greedy(
            delete_vertices(
                network,
                x
            )
        )
    }
)

which gives

[[1]]
IGRAPH clustering fast greedy, groups: 3, mod: 0.32
  groups:
  $`1`
  [1] 2 6 8 9

  $`2`
  [1] 1 5 7

  $`3`
  [1] 3 4


[[2]]
IGRAPH clustering fast greedy, groups: 3, mod: 0.29
  groups:
  $`1`
  [1] 5 7 8

  $`2`
  [1] 1 4 6

  $`3`
  [1] 2 3


[[3]]
IGRAPH clustering fast greedy, groups: 2, mod: 0.36
  groups:
  $`1`
  [1] 2 3 6 7

  $`2`
  [1] 1 4 5


[[4]]
IGRAPH clustering fast greedy, groups: 2, mod: 0.32
  groups:
  $`1`
  [1] 2 3 6

  $`2`
  [1] 1 4 5


[[5]]
IGRAPH clustering fast greedy, groups: 2, mod: 0.38
  groups:
  $`1`
  [1] 1 4 5

  $`2`
  [1] 2 3