Update example with spmatrix. (#5219)

* update

* graph
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>

examples/sparse/appnp.py

@@ -94,9 +94,9 @@ if __name__ == "__main__":
     g = dataset[0].to(dev)
 
     # Create the sparse adjacency matrix A.
-    src, dst = g.edges()
+    indices = torch.stack(g.edges())
     N = g.num_nodes()
-    A = dglsp.from_coo(dst, src, shape=(N, N))
+    A = dglsp.spmatrix(indices, shape=(N, N))
 
     # Calculate the symmetrically normalized adjacency matrix.
     I = dglsp.identity(A.shape, device=dev)

examples/sparse/c_and_s.py

@@ -99,9 +99,9 @@ if __name__ == "__main__":
     g = dataset[0].to(dev)
 
     # Create the sparse adjacency matrix A.
-    src, dst = g.edges()
+    indices = torch.stack(g.edges())
     N = g.num_nodes()
-    A = dglsp.from_coo(dst, src, shape=(N, N))
+    A = dglsp.spmatrix(indices, shape=(N, N))
 
     # Calculate the symmetrically normalized adjacency matrix.
     I = dglsp.identity(A.shape, device=dev)

examples/sparse/gat.py

@@ -123,9 +123,9 @@ if __name__ == "__main__":
     g = dataset[0].to(dev)
 
     # Create the sparse adjacency matrix A.
-    src, dst = g.edges()
+    indices = torch.stack(g.edges())
     N = g.num_nodes()
-    A = dglsp.from_coo(dst, src, shape=(N, N))
+    A = dglsp.spmatrix(indices, shape=(N, N))
 
     # Add self-loops.
     I = dglsp.identity(A.shape, device=dev)

examples/sparse/gcn.py

@@ -85,9 +85,9 @@ if __name__ == "__main__":
     X = g.ndata["feat"]
 
     # Create the adjacency matrix of graph.
-    src, dst = g.edges()
+    indices = torch.stack(g.edges())
     N = g.num_nodes()
-    A = dglsp.from_coo(dst, src, shape=(N, N))
+    A = dglsp.spmatrix(indices, shape=(N, N))
 
     ############################################################################
     # (HIGHLIGHT) Compute the symmetrically normalized adjacency matrix with

examples/sparse/gcnii.py

@@ -137,9 +137,9 @@ if __name__ == "__main__":
     H = g.ndata["feat"]
 
     # Create the adjacency matrix of graph.
-    src, dst = g.edges()
+    indices = torch.stack(g.edges())
     N = g.num_nodes()
-    A = dglsp.from_coo(dst, src, shape=(N, N))
+    A = dglsp.spmatrix(indices, shape=(N, N))
 
     ############################################################################
     # (HIGHLIGHT) Compute the symmetrically normalized adjacency matrix with

examples/sparse/graph_transformer.py

@@ -98,9 +98,9 @@ class GTModel(nn.Module):
         )
 
     def forward(self, g, X, pos_enc):
-        src, dst = g.edges()
+        indices = torch.stack(g.edges())
         N = g.num_nodes()
-        A = dglsp.from_coo(dst, src, shape=(N, N))
+        A = dglsp.spmatrix(indices, shape=(N, N))
         h = self.atom_encoder(X) + self.pos_linear(pos_enc)
         for layer in self.layers:
             h = layer(A, h)

examples/sparse/hgnn.py

@@ -71,8 +71,8 @@ def load_data():
     # self-loops).
     # We follow the paper and assume that the rows of the incidence matrix
     # are for nodes and the columns are for edges.
-    src, dst = graph.edges()
-    H = dglsp.from_coo(dst, src)
+    indices = torch.stack(graph.edges())
+    H = dglsp.spmatrix(indices)
     H = H + dglsp.identity(H.shape)
 
     X = graph.ndata["feat"]

examples/sparse/hypergraphatt.py

@@ -100,8 +100,8 @@ def load_data():
     # self-loops).
     # We follow the paper and assume that the rows of the incidence matrix
     # are for nodes and the columns are for edges.
-    src, dst = graph.edges()
-    H = dglsp.from_coo(dst, src)
+    indices = torch.stack(graph.edges())
+    H = dglsp.spmatrix(indices)
     H = H + dglsp.identity(H.shape)
 
     X = graph.ndata["feat"]

examples/sparse/sgc.py

@@ -69,9 +69,9 @@ if __name__ == "__main__":
     g = dataset[0].to(dev)
 
     # Create the sparse adjacency matrix A
-    src, dst = g.edges()
+    indices = torch.stack(g.edges())
     N = g.num_nodes()
-    A = dglsp.from_coo(dst, src, shape=(N, N))
+    A = dglsp.spmatrix(indices, shape=(N, N))
 
     # Calculate the symmetrically normalized adjacency matrix.
     I = dglsp.identity(A.shape, device=dev)

examples/sparse/sign.py

@@ -102,9 +102,9 @@ if __name__ == "__main__":
 
     # Create the sparse adjacency matrix A (note that W was used as the notation
     # for adjacency matrix in the original paper).
-    src, dst = g.edges()
+    indices = torch.stack(g.edges())
     N = g.num_nodes()
-    A = dglsp.from_coo(dst, src, shape=(N, N))
+    A = dglsp.spmatrix(indices, shape=(N, N))
 
     # Calculate the symmetrically normalized adjacency matrix.
     I = dglsp.identity(A.shape, device=dev)

examples/sparse/twirls.py

@@ -184,9 +184,9 @@ if __name__ == "__main__":
     X = g.ndata["feat"]
 
     # Create the sparse adjacency matrix A.
-    src, dst = g.edges()
+    indices = torch.stack(g.edges())
     N = g.num_nodes()
-    A = dglsp.from_coo(dst, src, shape=(N, N))
+    A = dglsp.spmatrix(indices, shape=(N, N))
 
     # Create the TWIRLS model.
     in_size = X.shape[1]