Adding edge weights to igraph from an adjacency matrix

Given a unipartite igraph object G and an adjacency matrix attrib containing an edge attribute for only some of the edges in G, what is the most efficient way to add the attributes to G.

Here's a reproducible example:

> set.seed(5)
> library(igraph)
> G <- erdos.renyi.game(10,.5,type="gnp")
> as_adjacency_matrix(G)  #Adjacency matrix for network

10 x 10 sparse Matrix of class "dgCMatrix"
                         
 [1,] . . . 1 1 . 1 1 1 1
 [2,] . . . 1 . . . 1 . .
 [3,] . . . 1 1 . 1 . 1 1
 [4,] 1 1 1 . 1 1 . . 1 1
 [5,] 1 . 1 1 . 1 . . . 1
 [6,] . . . 1 1 . . . 1 1
 [7,] 1 . 1 . . . . . 1 1
 [8,] 1 1 . . . . . . . .
 [9,] 1 . 1 1 . 1 1 . . .
[10,] 1 . 1 1 1 1 1 . . .

> attrib <- as_adjacency_matrix(G, sparse = FALSE)
> attrib[attrib==1] <- sample(c(0,1),sum(attrib),replace=TRUE)
> attrib  #Adjacency matrix containing attributes

      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
 [1,]    0    0    0    1    1    0    0    0    1     0
 [2,]    0    0    0    0    0    0    0    1    0     0
 [3,]    0    0    0    0    0    0    1    0    1     1
 [4,]    0    1    1    0    1    1    0    0    0     0
 [5,]    1    0    1    0    0    0    0    0    0     1
 [6,]    0    0    0    1    0    0    0    0    1     0
 [7,]    0    0    1    0    0    0    0    0    0     1
 [8,]    1    0    0    0    0    0    0    0    0     0
 [9,]    1    0    1    1    0    1    0    0    0     0
[10,]    1    0    0    0    0    0    1    0    0     0

One possibility is to make attrib an edgelist, then add them one-by-one in a loop. But, I'm thinking there must be a better option. Thanks for any ideas.

I guess this should work for you

G <- G %>%
    set_edge_attr("attrib", value = pmax(attrib,t(attrib))[as_edgelist(G)])

which gives

> G
IGRAPH c11c0d1 U--- 10 23 -- Erdos-Renyi (gnp) graph
+ attr: name (g/c), type (g/c), loops (g/l), p (g/n), attrib (e/n)
+ edges from c11c0d1:
 [1] 1-- 4 2-- 4 3-- 4 1-- 5 3-- 5 4-- 5 4-- 6 5-- 6 1-- 7 3-- 7 1-- 8 2-- 8
[13] 1-- 9 3-- 9 4-- 9 6-- 9 7-- 9 1--10 3--10 4--10 5--10 6--10 7--10

> E(G)$attrib
 [1] 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 0 1 1 0 1 0 1

and plot(G, edge.label = E(G)$attrib) shows

Input data

set.seed(5)
library(igraph)
n <- 10
G <- sample_gnp(n, 0.5, directed=FALSE)
G <- simplify(G, remove.multiple = TRUE, remove.loops = TRUE)

attrib <- as_adjacency_matrix(G, sparse = FALSE)
attrib[attrib == 1] <- sample(c(0, 1),sum(attrib),replace = TRUE)
attrib  #Adjacency matrix containing attributes

Solution using triplets (@i, @j, @x) in sparse matrix representation

# Using the Matrix package.
# Create sparse matrix (M).
# Create pairwise sequence of incident vertices (vp), x1, y1, x2, y2, ...
# Create a list of values (M@x)
library(Matrix)
M  <- as(attrib, "TsparseMatrix")
M  <- triu(as(pmax(attrib, t(attrib)), "TsparseMatrix")) # or save some space when undirected.
vp <- c(rbind(M@i, M@j)) + 1

# Convert vertex sequence to edge ids.
# Set edge attributes by edge id.
eids <- get.edge.ids(G, vp, directed=FALSE)
g2   <- set_edge_attr(G, "weight", index = eids, value=M@x)

# Show edges without weight.
E(g2)[which(is.na(E(g2)$weight))]
# + 5/23 edges from ad6a11e:
# [1] 1-- 7 4--10 5-- 6 6--10 7-- 9

g3 <- set_edge_attr(g2, name = "label", value = E(g2)$weight)
plot(g3)