Update tests

Co-Authored-By: Claudio Moroni <[email protected]>
Co-Authored-By: Pietro Monticone <[email protected]>

Author: 3 people

test/abstractmultilayerdigraph.jl

@@ -60,7 +60,7 @@ for node in nodes(multilayerdigraph)
     @test has_node(multilayerdigraph, node)
 end
 
-## Test MultilayerGraphs.add_node! and MultilayerGraphs.rem_node!
+# Test MultilayerGraphs.add_node! and MultilayerGraphs.rem_node!
 new_node = Node("new_node")
 nv_prev = nv(multilayerdigraph)
 ne_prev = ne(multilayerdigraph)
@@ -69,7 +69,7 @@ ne_prev = ne(multilayerdigraph)
 @test has_node(multilayerdigraph, new_node)
 @test MultilayerGraphs.rem_node!(multilayerdigraph, new_node)
 @test !has_node(multilayerdigraph, new_node)
-### Test that nothing changed
+# Test that nothing changed
 @test nv_prev == nv(multilayerdigraph)
 @test ne_prev == ne(multilayerdigraph)
 
@@ -78,19 +78,17 @@ ne_prev = ne(multilayerdigraph)
 nv(multilayerdigraph)
 @test length(multilayerdigraph.fadjlist) == length(vertices(multilayerdigraph)) # nv_withmissing(multilayerdigraph)
 
-## Test that all multilayer vertices are present
+# Test that all multilayer vertices are present
 for mv in vcat(mv_vertices.(all_layers_d)...)
     @test has_vertex(multilayerdigraph, mv)
 end
 
 for mv in mv_vertices(multilayerdigraph)
-    # if !(mv isa MissingVertex)
         mv_inneighbors(multilayerdigraph, mv)
         mv_outneighbors(multilayerdigraph, mv)
-    # end
 end
 
-## Test add_vertex! and rem_vertex!
+# Test add_vertex! and rem_vertex!
 
 # Test edges
 ne(multilayerdigraph)
@@ -116,7 +114,6 @@ wgt = weight_tensor(multilayerdigraph)
 wgt = weight_tensor(multilayerdigraph)
 @test wgt[rand_mv_1_weight, rand_mv_2_weight] == get_weight(multilayerdigraph, rand_mv_1_weight, rand_mv_2_weight) == _weight + 1
 
-
 ## Test set_metadata!
 _, rand_mv_1_meta, rand_mv_2_meta = _get_srcmv_dstmv_layer(layer_mdg)
 ### On vertices
@@ -135,7 +132,6 @@ mt = metadata_tensor(multilayerdigraph)
 
 # Test Graphs.jl extra overrides
 @test all(indegree(multilayerdigraph) .+ outdegree(multilayerdigraph) .== degree(multilayerdigraph))
-
 @inferred(mean_degree(multilayerdigraph))
 @inferred(degree_second_moment(multilayerdigraph))
 @inferred(degree_variance(multilayerdigraph))
@@ -154,8 +150,6 @@ eig_centr_u, errs_u = eigenvector_centrality(multilayerdigraph; norm="n", tol=1e
 
 modularity(multilayerdigraph, rand([1, 2, 3, 4], length(nodes(multilayerdigraph)), length(multilayerdigraph.layers)), )
 
-# get_graph_of_layers(multilayerdigraph)
-
 wgt = weight_tensor(multilayerdigraph)
 sam = supra_weight_matrix(multilayerdigraph)
 for edge in collect(edges(multilayerdigraph.layer_swdg))
@@ -167,33 +161,19 @@ end
 # Test that, given a 1-dimensional multilayerdigraph, we obtain the same metrics as we would by using Graphs.jl utilities on the one and only layer
 ## unweighted and weighted case
 for layer in all_layers_d
-
     if !(layer.graph isa SimpleValueGraphs.AbstractValGraph)
         monolayergraph = MultilayerDiGraph([layer])
-
         @test length(edges(monolayergraph)) == length(edges(layer.graph))
-
         @test eltype(monolayergraph) == eltype(layer.graph)
-
         @test ne(monolayergraph) == ne(layer.graph)
-
         @test length(nodes(monolayergraph)) == nv(layer.graph)
-
         @test nv(monolayergraph) .== nv(layer.graph)
-
         @test all(inneighbors.(Ref(monolayergraph), vertices(monolayergraph)) .== inneighbors.(Ref(layer.graph), vertices(layer.graph)))
-
-
         @test all(indegree(monolayergraph) .== indegree(layer.graph))
-
         @test all(outdegree(monolayergraph) .== outdegree(layer.graph))
-
         @test all(degree(monolayergraph) .== degree(layer.graph))
-
         @test_broken vec(eigenvector_centrality(monolayergraph; norm="n", tol=1e-3)[1]) ==
             eigenvector_centrality(layer.graph)
-
-
         tests = Bool[]
         for i in 1:5
             clustering = rand(
@@ -205,7 +185,6 @@ for layer in all_layers_d
             )
         end
         @test_broken all(tests)
-
         for edge in edges(layer)
             @test has_edge(monolayergraph, edge)
         end

test/abstractmultilayerugraph.jl

@@ -1,12 +1,8 @@
 all_layers_u = [layer for layer in all_layers if !is_directed(layer)]
 all_interlayers_u = [interlayer for interlayer in all_interlayers if !is_directed(interlayer)]
-
 multilayergraph = MultilayerGraph(all_layers_u, all_interlayers_u)
-
 layers_to_be_emptied =  deepcopy([layer for layer in all_layers_u if !(layer.graph isa SimpleWeightedGraphs.AbstractSimpleWeightedGraph)])
-
 layers_names_to_be_emptied = name.(layers_to_be_emptied)
-
 interlayers_to_be_emptied =  deepcopy([interlayer for interlayer in all_interlayers_u if all(in.(interlayer.layers_names, Ref(layers_names_to_be_emptied))) && !(interlayer.graph isa SimpleWeightedGraphs.AbstractSimpleWeightedGraph) ])
 
 for layer in layers_to_be_emptied
@@ -22,7 +18,6 @@ for interlayer in interlayers_to_be_emptied
 end
 
 @test all(ne.(layers_to_be_emptied) .== 0)
-
 @test all(ne.(interlayers_to_be_emptied) .== 0)
 
 # Instantiate configuration-model multilayergraph
@@ -58,7 +53,6 @@ default_edge_metadata = (src,dst) -> (rand(), "missing edge from_$(src)_to_$(dst
 
 # Test nodes
 @inferred(nodes(multilayergraph))
-
 @inferred(nn(multilayergraph))
 
 for node in nodes(multilayergraph)
@@ -90,7 +84,6 @@ for mv in vcat(mv_vertices.(all_layers_u)...)
 end
 
 for mv in mv_vertices(multilayergraph)
-
     mv_inneighbors(multilayergraph, mv)
     mv_outneighbors(multilayergraph, mv)
 end
@@ -147,14 +140,10 @@ mt = metadata_tensor(multilayergraph)
 @test all(MultilayerGraphs.get_supra_weight_matrix_from_weight_tensor(weight_tensor(multilayergraph).array) .== supra_weight_matrix(multilayergraph).array)
 @test all(MultilayerGraphs.get_weight_tensor_from_supra_weight_matrix(multilayergraph, supra_weight_matrix(multilayergraph).array) .==  weight_tensor(multilayergraph).array)
 @test_broken multilayer_global_clustering_coefficient(multilayergraph) .== global_clustering_coefficient(multilayergraph)
-
 overlaygraph = MultilayerGraphs.get_overlay_monoplex_graph(multilayergraph)
 @test_broken global_clustering_coefficient(overlaygraph) .== overlay_clustering_coefficient(multilayergraph)
-
 @test multilayer_weighted_global_clustering_coefficient(multilayergraph, [1 / 3, 1 / 3, 1 / 3]) .≈ multilayer_global_clustering_coefficient(multilayergraph)
-
 eig_centr_u, errs_u = eigenvector_centrality(multilayergraph; norm="n", tol=1e-3)
-
 modularity(multilayergraph, rand([1, 2, 3, 4], length(nodes(multilayergraph)), length(multilayergraph.layers)), )
 
 # get_graph_of_layers(multilayergraph)
@@ -173,28 +162,17 @@ for layer in all_layers_u
 
     if !(layer.graph isa SimpleValueGraphs.AbstractValGraph)
         monolayergraph = MultilayerGraph([layer])
-
         @test length(edges(monolayergraph)) == length(edges(layer.graph))
-
         @test eltype(monolayergraph) == eltype(layer.graph)
-
         @test ne(monolayergraph) == ne(layer.graph)
-
         @test length(nodes(monolayergraph)) == nv(layer.graph)
-
         @test nv(monolayergraph) .== nv(layer.graph)
-
         @test all(inneighbors.(Ref(monolayergraph), vertices(monolayergraph)) .== inneighbors.(Ref(layer.graph), vertices(layer.graph)))
-
         @test all(indegree(monolayergraph) .== indegree(layer.graph))
-
         @test all(outdegree(monolayergraph) .== outdegree(layer.graph))
-
         @test all(degree(monolayergraph) .== degree(layer.graph))
-
         @test_broken vec(eigenvector_centrality(monolayergraph; norm="n", tol=1e-3)[1]) ==
             eigenvector_centrality(layer.graph)
-
         tests = Bool[]
         for i in 1:5
             clustering = rand(
@@ -206,7 +184,6 @@ for layer in all_layers_u
             )
         end
         @test_broken all(tests)
-
         for edge in edges(layer)
             @test has_edge(monolayergraph, edge)
         end

test/agents_jl_integration.jl

@@ -1,36 +1,38 @@
-graph = MultilayerGraph(all_layers_u, all_interlayers_u)
+################################################
+####### TEST INTEGRATION WITH AGENTS.JL ########
+################################################
 
+# Create a multilayer graph 
+graph = MultilayerGraph(all_layers_u, all_interlayers_u)
 # Define the agent type
 @agent EigenAgent GraphAgent begin
     previous_value::Float64
     old_value::Float64
     new_value::Float64
 end
-
+# Instantiate the agent
 EigenAgent(id, pos; initial_opinion::Float64) = EigenAgent(id, pos, initial_opinion, initial_opinion,-1.0)
-
-# Define the ABM
-EigenABM = ABM(EigenAgent, GraphSpace(graph)) # 
-
-# Add agents so that agent i is in vertex i
+# Define the agent-based model
+EigenABM = ABM(EigenAgent, GraphSpace(graph)) 
+# Add agents to the ABM so that agent i is located in vertex i
 for (i,mv) in enumerate(mv_vertices(graph))
     initial_opinion = 1.0
     add_agent!(i,EigenABM, initial_opinion = initial_opinion)
 end
-
+# Define the individual-level dynamics (micro-dynamics)
 function agent_step!(agent::EigenAgent,model)
     agent.previous_value = agent.old_value
     agent.new_value =  sum([outneighbor_agent.old_value for outneighbor_agent in nearby_agents(agent, model; neighbor_type = :all)])
 end
-
+# Define the system-level dynamics (macro-dynamics)
 function model_step!(model)
-    tot_edges_weight = sqrt(sum(abs2, [agent.new_value for agent in allagents(model)])) #sum(weight_tensor(model.space.graph).array)
+    tot_edges_weight = sqrt(sum(abs2, [agent.new_value for agent in allagents(model)]))
     for agent in allagents(model)
         agent.new_value = agent.new_value / tot_edges_weight
         agent.old_value = agent.new_value
     end
 end
-
+# Set the rule to stop the model simulation  
 function terminate(model, s)
     # println(s, typeof(s))
     if any(
@@ -42,13 +44,11 @@ function terminate(model, s)
         return true
     end
 end
-
+# Simulate the model 
 agent_data, _ = run!(EigenABM, agent_step!, model_step!, terminate; adata = [:new_value], when = terminate)
-
+# Compute the eigenvector centrality of the surrounding multilayer graph 
 eig_centr_swm, err_swm =  eigenvector_centrality(EigenABM.space.graph ;weighted = false,  tol = 1e-18, norm = "null")
 
 for (val_1, val_2) in zip(eig_centr_swm, agent_data.new_value)
     @test val_1 ≈ val_2
-end
-
-# @test all(eig_centr_swm .≈ agent_data.new_value)
+end