from torch_geometric.datasets import Planetoid from torch_geometric.utils import to_scipy_sparse_matrix dataset = Planetoid(root='', name='Cora') # 将数据保存在data文件夹下 data = dataset[0] adj = to_scipy_sparse_matrix(data.edge_index) #将数据变成邻接矩阵的形式 features = data.x labels ...
# get adj adj = to_scipy_sparse_matrix(edge_index).todense() adj = torch.tensor(adj).to(device) 提取度矩阵: deg = degree(edge_index[0], dataset.num_nodes) deg = torch.diag_embed(deg) deg_inv_sqrt = torch.pow(deg, -0.5) deg_inv_sqrt[deg_inv_sqrt == float('inf')] = 0 ...
Fixed bug in to_scipy_sparse_matrix() when CUDA is set as default torch device (#9146) Fixed the MetaPath2Vec model in case the last node is isolated (#9145) Ensured backward compatibility in MessagePassing via torch.load() (#9105) Prevented model compilation on custom MessagePassing.propag...
adj = nx.to_scipy_sparse_array(G).tocoo() File "F:\anaconda\envs\ai\lib\site-packages\networkx\convert_matrix.py", line 921, in to_scipy_sparse_array A = sp.sparse.coo_array((d, (r, c)), shape=(nlen, nlen), dtype=dtype) AttributeError: module 'scipy.sparse' has no attribu...
matmul(x) # Sparse-dense matrix multiplication adj = adj.matmul(adj) # Sparse-sparse matrix multiplication # Creating SparseTensor instances: adj = SparseTensor.from_dense(mat) adj = SparseTensor.eye(100, 100) adj = SparseTensor.from_scipy(mat) MessagePassing接口可以使用Tensor类或新的Sparse...
dense_to_sparse, one_hot, to_dense_adj, to_scipy_sparse_matrix, )class UnpoolInfo(NamedTuple): edge_index: Tensor cluster: Tensor batch: Tensorclass ClusterPooling(torch.nn.Module): r"""The cluster pooling operator from the `"Edge-Based Graph Component Pooling" <paper url>`_ paper.:...
def matrix_U(N): u = lambda n, N: np.cos(2 * np.pi / N * n * np.arange(N)) - 1j * np.sin(2 * np.pi / N * n * np.arange(N)) U = np.empty((N, 0)) for n in range(N): U = np.c_[U, u(n, N)] return U def fourier_transform(v): N = v.shape[0]...
You can create a scipy matrix for each individual attention head: adjs = [] for i in range(attention_score.size(1)): adj = torch_geometric.utils.to_scipy_sparse_matrix(indices, attention_score[:, i]) adjs.append(adj) 👍 4 pinkfloyd06 commented Jun 25, 2021 Thanks, it helps ...
data.pos: Node position matrix with shape[num_nodes, num_dimensions] 一个简单的例子:(无向图以双向存储) importtorchfromtorch_geometric.dataimportDataedge_index=torch.tensor([[0,1,1,2],[1,0,2,1]],dtype=torch.long)x=torch.tensor([[-1],[0],[1]],dtype=torch.float)data=Data(x=x,ed...
G = sp.coo_matrix(W) edge_index = torch.tensor(np.array([G.row, G.col]), dtype=torch.int64).to(device) edge_weight =torch.tensor(G.data).float().to(device) 训练 首先初始化模型, 定义损失函数和优化器. # 构建模型 model = TrafficModel(device, num_nodes, channels, num_layers, kern...