def test(adj): ''' test on GCN ''' # adj = normalize_adj_tensor(adj) gcn = GCN(nfeat=features.shape[1], nhid=16, nclass=labels.max().item() + 1, dropout=0.5, device=device) gcn = gcn.to(device) optimizer = optim.Adam(gcn.parameters(), lr=0.01, weight_decay=5e-4) gcn.fit(features, adj, labels, idx_train) # train without model picking # gcn.fit(features, adj, labels, idx_train, idx_val) # train with validation model picking output = gcn.output loss_test = F.nll_loss(output[idx_test], labels[idx_test]) acc_test = accuracy(output[idx_test], labels[idx_test]) print("Test set results:", "loss= {:.4f}".format(loss_test.item()), "accuracy= {:.4f}".format(acc_test.item())) return acc_test.item()
def test_gcn(adj, data, cuda, data_prep, nhid=16): ''' test on GCN ''' device = torch.device("cuda" if cuda else "cpu") features, labels, idx_train, idx_val, idx_test = data_prep(data, device) gcn = GCN(nfeat=features.shape[1], nhid=nhid, nclass=labels.max().item() + 1, dropout=0.5, device=device) gcn = gcn.to(device) optimizer = optim.Adam(gcn.parameters(), lr=0.01, weight_decay=5e-4) gcn.fit(features, adj, labels, idx_train) # train without model picking # gcn.fit(features, adj, labels, idx_train, idx_val) # train with validation model picking output = gcn.output loss_test = F.nll_loss(output[idx_test], labels[idx_test]) acc_test = accuracy(output[idx_test], labels[idx_test]) return acc_test.item()