The edge weights can be of type Bool, Int8, UInt8, Int16, UInt16, Int32, UInt32, Int64, UInt64, Float32 or Float64 (i.e., the GraphBLAS predefined types). User-defined types are not supported.
julia> using GraphBLASInterface, SuiteSparseGraphBLAS, LightGraphsGraphBLAS, LightGraphs
julia> GrB_init(GrB_NONBLOCKING)
GrB_SUCCESS::GrB_Info = 0
julia> g = BLASGraph{Int64}(5) # edge weights are of type Int64, the eltype for all graphs is UInt64 and cannot be changed
{5, 0} undirected graph
julia> add_edge!(g, 2, 4, 5)
true
julia> add_edge!(g, 1, 2, 7)
true
julia> foreach(println, edges(g))
Edge 1 => 2 with weight 7
Edge 2 => 4 with weight 5
julia> rem_edge!(g, 2, 4)
true
julia> foreach(println, edges(g))
Edge 1 => 2 with weight 7
julia> add_edge!(g, 1, 2, 9) # reweight the edge
true
julia> foreach(println, edges(g))
Edge 1 => 2 with weight 9Create graph from AbstractSimpleGraph :
julia> g = BLASDiGraph(SimpleDiGraph(5, 6))
{5, 6} directed graph
julia> foreach(println, edges(g))
Edge 1 => 3 with weight 1
Edge 1 => 4 with weight 1
Edge 1 => 5 with weight 1
Edge 2 => 5 with weight 1
Edge 3 => 5 with weight 1
Edge 4 => 2 with weight 1Create graph from GraphBLAS matrix :
julia> I = OneBasedIndex[1, 2, 3]; J = OneBasedIndex[2, 3, 1]; X = Float64[7.2, 3.4, 5.6];
julia> M = GrB_Matrix(I, J, X)
GrB_Matrix{Float64}
julia> g = BLASDiGraph(M)
{3, 3} directed graph
julia> foreach(println, edges(g))
Edge 1 => 2 with weight 7.2
Edge 2 => 3 with weight 3.4
Edge 3 => 1 with weight 5.6
julia> outneighbors(g, 1) # vertices of the graph are always of type UInt64 since they are indices of a GraphBLAS matrix
1-element Array{UInt64,1}:
0x0000000000000002
julia> inneighbors(g, 1)
1-element Array{UInt64,1}:
0x0000000000000003Create a graph from edge list :
julia> e1 = SimpleWeightedEdge(1, 2, Int32(4))
Edge 1 => 2 with weight 4
julia> e2 = SimpleWeightedEdge(2, 4, Int32(8))
Edge 2 => 4 with weight 8
julia> edge_list = [e1, e2]
2-element Array{SimpleWeightedEdge{Int64,Int32},1}:
Edge 1 => 2 with weight 4
Edge 2 => 4 with weight 8
julia> g = BLASGraph(edge_list)
{4, 2} undirected graph
julia> get_weight(g, 1, 2)
4Create a graph from adjacency matrix :
julia> A = [0 1 1; 1 0 0; 0 1 0]
3×3 Array{Int64,2}:
0 1 1
1 0 0
0 1 0
julia> g = BLASDiGraph(A)
{3, 4} directed graph
julia> indegree(g, 1)
1
julia> outdegree(g, 1)
2
julia> using SparseArrays
julia> B = sparse(A)
3×3 SparseMatrixCSC{Int64,Int64} with 4 stored entries:
[2, 1] = 1
[1, 2] = 1
[3, 2] = 1
[1, 3] = 1
julia> g = BLASDiGraph(B)
{3, 4} directed graphOn 6 cores -
gdistances (algorithm: bfs_simple) :
julia> using GraphBLASInterface, SuiteSparseGraphBLAS, LightGraphsGraphBLAS, LightGraphs, MatrixDepot, BenchmarkTools, SNAPDatasets
julia> md = mdopen("DIMACS10/caidaRouterLevel");
julia> A = md.A
julia> lg = SimpleGraph(A)
{192244, 609066} undirected simple Int64 graph
julia> GrB_init(GrB_NONBLOCKING)
GrB_SUCCESS::GrB_Info = 0
julia> bg = BLASGraph(lg)
{192244, 609066} undirected graph
julia> @benchmark gdistances(lg, source) setup = (source = rand(1:nv(lg)))
BenchmarkTools.Trial:
memory estimate: 4.42 MiB
allocs estimate: 9
--------------
minimum time: 159.524 μs (0.00% GC)
median time: 29.321 ms (0.00% GC)
mean time: 29.411 ms (0.00% GC)
maximum time: 34.208 ms (0.00% GC)
--------------
samples: 170
evals/sample: 1
julia> @benchmark gdistances(bg, source) setup = (source = rand(1:nv(bg)))
BenchmarkTools.Trial:
memory estimate: 2.55 KiB
allocs estimate: 95
--------------
minimum time: 42.370 ms (0.00% GC)
median time: 45.216 ms (0.00% GC)
mean time: 45.386 ms (0.00% GC)
maximum time: 53.582 ms (0.00% GC)
--------------
samples: 111
evals/sample: 1
julia> lg = loadsnap(:soc_slashdot0902_u)
{82168, 582533} undirected simple Int64 graph
julia> bg = BLASGraph(lg)
{82168, 582533} undirected graph
julia> @benchmark gdistances(lg, source) setup = (source = rand(1:nv(lg)))
BenchmarkTools.Trial:
memory estimate: 1.89 MiB
allocs estimate: 8
--------------
minimum time: 9.511 ms (0.00% GC)
median time: 10.535 ms (0.00% GC)
mean time: 10.542 ms (0.20% GC)
maximum time: 12.301 ms (0.00% GC)
--------------
samples: 473
evals/sample: 1
julia> @benchmark gdistances(bg, source) setup = (source = rand(1:nv(bg)))
BenchmarkTools.Trial:
memory estimate: 1.56 KiB
allocs estimate: 60
--------------
minimum time: 16.929 ms (0.00% GC)
median time: 20.320 ms (0.00% GC)
mean time: 20.280 ms (0.00% GC)
maximum time: 35.234 ms (0.00% GC)
--------------
samples: 247
evals/sample: 1
julia> lg = loadsnap(:facebook_combined)
{4039, 88234} undirected simple Int64 graph
julia> bg = BLASGraph(lg)
{4039, 88234} undirected graph
julia> @benchmark gdistances(lg, source) setup = (source = rand(1:nv(lg)))
BenchmarkTools.Trial:
memory estimate: 95.69 KiB
allocs estimate: 8
--------------
minimum time: 425.521 μs (0.00% GC)
median time: 429.602 μs (0.00% GC)
mean time: 442.115 μs (0.29% GC)
maximum time: 1.709 ms (62.72% GC)
--------------
samples: 10000
evals/sample: 1
julia> @benchmark gdistances(bg, source) setup = (source = rand(1:nv(bg)))
BenchmarkTools.Trial:
memory estimate: 1.13 KiB
allocs estimate: 44
--------------
minimum time: 662.463 μs (0.00% GC)
median time: 1.472 ms (0.00% GC)
mean time: 1.478 ms (0.00% GC)
maximum time: 10.979 ms (0.00% GC)
--------------
samples: 3313
evals/sample: 1