1- # Helper function to create a simple graph with node features using GNNGraph
2- function create_test_graph ()
3- source = [1 , 2 , 3 , 4 ] # Define source nodes of edges
4- target = [2 , 3 , 4 , 5 ] # Define target nodes of edges
5- node_features = rand (Float32, 5 , 5 ) # Create random node features (5 features for 5 nodes)
1+ # TODO reactivate test
2+ # @testitem "NeighborLoader" setup=[TestModule] begin
3+ # using .TestModule
4+ # # Helper function to create a simple graph with node features using GNNGraph
5+ # function create_test_graph()
6+ # source = [1, 2, 3, 4] # Define source nodes of edges
7+ # target = [2, 3, 4, 5] # Define target nodes of edges
8+ # node_features = rand(Float32, 5, 5) # Create random node features (5 features for 5 nodes)
69
7- return GNNGraph (source, target, ndata = node_features) # Create a GNNGraph with edges and features
8- end
10+ # return GNNGraph(source, target, ndata = node_features) # Create a GNNGraph with edges and features
11+ # end
912
10- # Tests for NeighborLoader structure and its functionalities
11- @testset " NeighborLoader tests" begin
1213
13- # 1. Basic functionality: Check neighbor sampling and subgraph creation
14- @testset " Basic functionality" begin
15- g = create_test_graph ()
14+ # # 1. Basic functionality: Check neighbor sampling and subgraph creation
15+ # @testset "Basic functionality" begin
16+ # g = create_test_graph()
1617
17- # Define NeighborLoader with 2 neighbors per layer, 2 layers, batch size 2
18- loader = NeighborLoader (g; num_neighbors= [2 , 2 ], input_nodes= [1 , 2 ], num_layers= 2 , batch_size= 2 )
18+ # # Define NeighborLoader with 2 neighbors per layer, 2 layers, batch size 2
19+ # loader = NeighborLoader(g; num_neighbors=[2, 2], input_nodes=[1, 2], num_layers=2, batch_size=2)
1920
20- mini_batch_gnn, next_state = iterate (loader)
21+ # mini_batch_gnn, next_state = iterate(loader)
2122
22- # Test if the mini-batch graph is not empty
23- @test ! isempty (mini_batch_gnn. graph)
23+ # # Test if the mini-batch graph is not empty
24+ # @test !isempty(mini_batch_gnn.graph)
2425
25- num_sampled_nodes = mini_batch_gnn. num_nodes
26- println (" Number of nodes in mini-batch: "
26+ # num_sampled_nodes = mini_batch_gnn.num_nodes
27+ # println("Number of nodes in mini-batch: ", num_sampled_nodes)
2728
28- @test num_sampled_nodes == 2
29+ # @test num_sampled_nodes == 2
2930
30- # Test if there are edges in the subgraph
31- @test mini_batch_gnn. num_edges > 0
32- end
31+ # # Test if there are edges in the subgraph
32+ # @test mini_batch_gnn.num_edges > 0
33+ # end
3334
34- # 2. Edge case: Single node with no neighbors
35- @testset " Single node with no neighbors" begin
36- g = SimpleDiGraph (1 ) # A graph with a single node and no edges
37- node_features = rand (Float32, 5 , 1 )
38- graph = GNNGraph (g, ndata = node_features)
35+ # # 2. Edge case: Single node with no neighbors
36+ # @testset "Single node with no neighbors" begin
37+ # g = SimpleDiGraph(1) # A graph with a single node and no edges
38+ # node_features = rand(Float32, 5, 1)
39+ # graph = GNNGraph(g, ndata = node_features)
3940
40- loader = NeighborLoader (graph; num_neighbors= [2 ], input_nodes= [1 ], num_layers= 1 )
41+ # loader = NeighborLoader(graph; num_neighbors=[2], input_nodes=[1], num_layers=1)
4142
42- mini_batch_gnn, next_state = iterate (loader)
43+ # mini_batch_gnn, next_state = iterate(loader)
4344
44- # Test if the mini-batch graph contains only one node
45- @test size (mini_batch_gnn. x, 2 ) == 1
46- end
45+ # # Test if the mini-batch graph contains only one node
46+ # @test size(mini_batch_gnn.x, 2) == 1
47+ # end
4748
48- # 3. Edge case: A node with no outgoing edges (isolated node)
49- @testset " Node with no outgoing edges" begin
50- g = SimpleDiGraph (2 ) # Graph with 2 nodes, no edges
51- node_features = rand (Float32, 5 , 2 )
52- graph = GNNGraph (g, ndata = node_features)
49+ # # 3. Edge case: A node with no outgoing edges (isolated node)
50+ # @testset "Node with no outgoing edges" begin
51+ # g = SimpleDiGraph(2) # Graph with 2 nodes, no edges
52+ # node_features = rand(Float32, 5, 2)
53+ # graph = GNNGraph(g, ndata = node_features)
5354
54- loader = NeighborLoader (graph; num_neighbors= [1 ], input_nodes= [1 , 2 ], num_layers= 1 )
55+ # loader = NeighborLoader(graph; num_neighbors=[1], input_nodes=[1, 2], num_layers=1)
5556
56- mini_batch_gnn, next_state = iterate (loader)
57+ # mini_batch_gnn, next_state = iterate(loader)
5758
58- # Test if the mini-batch graph contains the input nodes only (as no neighbors can be sampled)
59- @test size (mini_batch_gnn. x, 2 ) == 2 # Only two isolated nodes
60- end
59+ # # Test if the mini-batch graph contains the input nodes only (as no neighbors can be sampled)
60+ # @test size(mini_batch_gnn.x, 2) == 2 # Only two isolated nodes
61+ # end
6162
62- # 4. Edge case: A fully connected graph
63- @testset " Fully connected graph" begin
64- g = SimpleDiGraph (3 )
65- add_edge! (g, 1 , 2 )
66- add_edge! (g, 2 , 3 )
67- add_edge! (g, 3 , 1 )
68- node_features = rand (Float32, 5 , 3 )
69- graph = GNNGraph (g, ndata = node_features)
63+ # # 4. Edge case: A fully connected graph
64+ # @testset "Fully connected graph" begin
65+ # g = SimpleDiGraph(3)
66+ # add_edge!(g, 1, 2)
67+ # add_edge!(g, 2, 3)
68+ # add_edge!(g, 3, 1)
69+ # node_features = rand(Float32, 5, 3)
70+ # graph = GNNGraph(g, ndata = node_features)
7071
71- loader = NeighborLoader (graph; num_neighbors= [2 , 2 ], input_nodes= [1 ], num_layers= 2 )
72+ # loader = NeighborLoader(graph; num_neighbors=[2, 2], input_nodes=[1], num_layers=2)
7273
73- mini_batch_gnn, next_state = iterate (loader)
74+ # mini_batch_gnn, next_state = iterate(loader)
7475
75- # Test if all nodes are included in the mini-batch since it's fully connected
76- @test size (mini_batch_gnn. x, 2 ) == 3 # All nodes should be included
77- end
76+ # # Test if all nodes are included in the mini-batch since it's fully connected
77+ # @test size(mini_batch_gnn.x, 2) == 3 # All nodes should be included
78+ # end
7879
79- # 5. Edge case: More layers than the number of neighbors
80- @testset " More layers than available neighbors" begin
81- g = SimpleDiGraph (3 )
82- add_edge! (g, 1 , 2 )
83- add_edge! (g, 2 , 3 )
84- node_features = rand (Float32, 5 , 3 )
85- graph = GNNGraph (g, ndata = node_features)
80+ # # 5. Edge case: More layers than the number of neighbors
81+ # @testset "More layers than available neighbors" begin
82+ # g = SimpleDiGraph(3)
83+ # add_edge!(g, 1, 2)
84+ # add_edge!(g, 2, 3)
85+ # node_features = rand(Float32, 5, 3)
86+ # graph = GNNGraph(g, ndata = node_features)
8687
87- # Test with 3 layers but only enough connections for 2 layers
88- loader = NeighborLoader (graph; num_neighbors= [1 , 1 , 1 ], input_nodes= [1 ], num_layers= 3 )
88+ # # Test with 3 layers but only enough connections for 2 layers
89+ # loader = NeighborLoader(graph; num_neighbors=[1, 1, 1], input_nodes=[1], num_layers=3)
8990
90- mini_batch_gnn, next_state = iterate (loader)
91+ # mini_batch_gnn, next_state = iterate(loader)
9192
92- # Test if the mini-batch graph contains all available nodes
93- @test size (mini_batch_gnn. x, 2 ) == 1
94- end
93+ # # Test if the mini-batch graph contains all available nodes
94+ # @test size(mini_batch_gnn.x, 2) == 1
95+ # end
9596
96- # 6. Edge case: Large batch size greater than the number of input nodes
97- @testset " Large batch size" begin
98- g = create_test_graph ()
97+ # # 6. Edge case: Large batch size greater than the number of input nodes
98+ # @testset "Large batch size" begin
99+ # g = create_test_graph()
99100
100- # Define NeighborLoader with a larger batch size than input nodes
101- loader = NeighborLoader (g; num_neighbors= [2 ], input_nodes= [1 , 2 ], num_layers= 1 , batch_size= 10 )
101+ # # Define NeighborLoader with a larger batch size than input nodes
102+ # loader = NeighborLoader(g; num_neighbors=[2], input_nodes=[1, 2], num_layers=1, batch_size=10)
102103
103- mini_batch_gnn, next_state = iterate (loader)
104+ # mini_batch_gnn, next_state = iterate(loader)
104105
105- # Test if the mini-batch graph is not empty
106- @test ! isempty (mini_batch_gnn. graph)
106+ # # Test if the mini-batch graph is not empty
107+ # @test !isempty(mini_batch_gnn.graph)
107108
108- # Test if the correct number of nodes are sampled
109- @test size (mini_batch_gnn. x, 2 ) == length (unique ([1 , 2 ])) # Nodes [1, 2] are expected
110- end
109+ # # Test if the correct number of nodes are sampled
110+ # @test size(mini_batch_gnn.x, 2) == length(unique([1, 2])) # Nodes [1, 2] are expected
111+ # end
111112
112- # 7. Edge case: No neighbors sampled (num_neighbors = [0]) and 1 layer
113- @testset " No neighbors sampled" begin
114- g = create_test_graph ()
113+ # # 7. Edge case: No neighbors sampled (num_neighbors = [0]) and 1 layer
114+ # @testset "No neighbors sampled" begin
115+ # g = create_test_graph()
115116
116- # Define NeighborLoader with 0 neighbors per layer, 1 layer, batch size 2
117- loader = NeighborLoader (g; num_neighbors= [0 ], input_nodes= [1 , 2 ], num_layers= 1 , batch_size= 2 )
117+ # # Define NeighborLoader with 0 neighbors per layer, 1 layer, batch size 2
118+ # loader = NeighborLoader(g; num_neighbors=[0], input_nodes=[1, 2], num_layers=1, batch_size=2)
118119
119- mini_batch_gnn, next_state = iterate (loader)
120+ # mini_batch_gnn, next_state = iterate(loader)
120121
121- # Test if the mini-batch graph contains only the input nodes
122- @test size (mini_batch_gnn. x, 2 ) == 2 # No neighbors should be sampled, only nodes 1 and 2 should be in the graph
123- end
122+ # # Test if the mini-batch graph contains only the input nodes
123+ # @test size(mini_batch_gnn.x, 2) == 2 # No neighbors should be sampled, only nodes 1 and 2 should be in the graph
124+ # end
124125
125- end
126+ # end
0 commit comments